CogLingDays 6 - Contragram research network

Faculty of Arts and Philosophy
CogLingDays 6
Biannual meeting of the Belgium Netherlands
Cognitive Linguistics Association (BeNeCLA)
BOOK OF ABSTRACTS
Ghent
11th and 12th December 2014
Organising committee
Frank Brisard (University of Antwerp)
Timothy Colleman (Ghent University)
Renata Enghels (Ghent University)
Tim Geleyn (Ghent University)
Peter Lauwers (Ghent University)
Paul Sambre (University of Leuven)
Miriam Taverniers (Ghent University)
Scientific committee
Jóhanna Barđdal (Ghent)
Ronny Boogaart (Leiden)
Frank Brisard (Antwerp)
Bert Cappelle (Lille)
Alan Cienki (Amsterdam)
Timothy Colleman (Ghent)
Bert Cornillie (Leuven)
Hubert Cuyckens (Leuven)
Barbara De Cock (Louvain)
Gert De Sutter (Ghent)
Renata Enghels (Ghent)
Kurt Feyaerts (Leuven)
Ad Foolen (Nijmegen)
Liesbet Heyvaert (Leuven)
Francisco Gonzálvez-García (Almeria)
Peter Lauwers (Ghent)
Dominique Legallois (Caen)
Maarten Lemmens (Lille)
Tanja Mortelmans (Antwerp)
Esther Pascual (Groningen)
Paul Sambre (Leuven)
Miriam Taverniers (Ghent)
Arie Verhagen (Leiden)
Dominique Willems (Ghent)
Table of Contents
Plenary Papers ...................................................................................................................... 3
Theme Session Papers: Cognitive Linguistics and Phenomenology ................................. 11
Regular Session Papers....................................................................................................... 21
1
Plenary Papers
3
Speakers, usage, culture: language as a phenomenon of the
third kind
Ewa Dąbrowska (Northumbria University)
Why are languages the way they are? Cognitive theories attempt to explain language
structure by appealing to cognitive mechanisms, which may include general abilities
such as categorization and pattern-finding skills, innate linguistic representations, or
some combination of the two; functional theories, on the other hand, emphasize
discourse pressures and processing limitations. While it is undeniable that both of these
play an important role, many aspects of grammar are best understood by appealing to
cultural processes such as language transmission, community size, exotericity,
standarization and education. I argue that higher-level patterns in language are not
necessarily represented in individual speakers’ mental grammars, and that community
grammars are typically more regular than the grammars of individual speakers.
Languages belong to communities, and individual speakers “own” only parts of their
language. To explain language structure, it is important to distingush between
individual grammars, the patterns found in the language as a whole, and the complex
interactions between them.
5
When the Ground turns into an abyss
Dirk Geeraerts (University of Leuven)
If empathy is the ability to see another person's point of view, then the ability to see
another person's conceptualization as incorporating a model of one's own point of view
may be called 'second order empathy'. In this talk I will explore how first and second
order empathy have an impact on the notion of Ground as used in Cognitive Linguistics.
Specifically, I will argue, first, thatempathetic construal points to the necessity of
incorporating the interlocutor as an independent conceptualizer into an intersubjective
model of the Ground, and second, that the possibility of choosing between first order
empathetic and second order empathetic readings of an interlocutor's utterances may
introduce an irreducible interpretative aporia in the Ground.
6
On parents and peers in constructional networks
Muriel Norde (Humboldt University Berlin)
Linguistic networks are a fundamental concept in construction grammars of various
kinds (e.g. Croft 2001, Goldberg 2006, Bybee 2010, Traugott & Trousdale 2013), but also in
other frameworks such as Word Grammar (Hudson 2007). Drawing on this (rapidly
expanding) body of research, I will discuss the types of links that can be identified in
networks of morphologically related nodes. I will argue that there are two main types
of links: inheritance links (the “parents“), which form hierarchical structure, and lateral
links (the “peers”) which link nodes on the same level of schematicity. Inheritance links
relate constructional schemas (e.g. word formation schemas or inflectional schemas) to
their instantiations at the micro-constructional level, whereby a micro-construction
may inherit from multiple schemas (cf. Goldberg 1995). Within the category of lateral
links, I will furthermore distinguish between (i) intraparadigmatic links (adapted from
Booij’s (2010) paradigmatic links), i.e. micro-constructions with the same base (for
either inflection or derivation), (ii) interparadigmatic links, i.e. micro-constructions that
inherit from the same schema(s), and (iii) resemblance links, linking phonologically and
/ or semantically similar micro-constructions that are not paradigmatically related. In
addition, lateral links can be observed to exist on the level of (sub-)schemas, a
phenomenon termed degeneracy (Van de Velde 2014).
Special emphasis will be on morphological change, which will be recast as
realignment in parts of the network. More specifically, I will explore the varying roles of
parents and peers in case of massive constructional realignment (such as the loss of
schemas in deflexion, or the emergence of new (sub-)schemas). Using data from
particular instances of change in Swedish and Dutch (Norde 2008, Norde & Morris 2013,
Norde & Trousdale ms.), I aim to show that these changes are constrained by two basic
constructional properties: connectivity, i.e. the number of links which connect microconstructions to their (multiple) parents and peers, and frequency, both at the microconstructional level (i.e. token frequency) and at the schema level (i.e. type frequency).
References
Booij, Geert. 2010. Construction Morphology. Oxford: Oxford University
Bybee; Joan, 2010. Language, usage and cognition. Cambridge: Cambridge University Press.
Croft, William. 2001. Radical Construction Grammar: syntactic theory in typological perspective.
Oxford: Oxford University Press.
Goldberg. 1995. Constructions. A Construction Grammar approach to argument structure. Chicago:
University of Chicago Press.
Goldberg, Adele. 2006. Constructions at work: the nature of generalization in language. Oxford: Oxford
University Press.
7
Hudson, Richard. 2007. Language networks. The new Word Grammar. Oxford: Oxford University
Press.
Norde, Muriel. 2008. East Scandinavian pd; a preposition of the ‘third kind’. In Kees Dekker,
Alasdair MacDonald & Hermann Niebaum (eds) Northern voices. Essays on Old Germanic
and related topics ofiered to Professor Tette Hofstra (= Germania Latina VI), 383-398. Leuven:
Peeters.
Norde, Muriel & Caroline Morris. 2014. Category change and constructional schemas: the case of
diminutive intensifying prefixoids in Dutch. Paper presented at ICCG 8, University of
Osnabrück, September 5, 2014.
Norde, Muriel & Graeme Trousdale (ms.). Exaptation and constructional change.
Traugott, Elizabeth Closs & Graeme Trousdale. 2013. Constructionalization and constructional
changes. Oxford: Oxford University Press.
Van de Velde, Freek. 2014. Degeneracy: The maintenance of constructional networks. In
Boogaart, Ronny, Timothy Colleman & Gijsbert Rutten (eds) Extending the scope of
construction grammar, 141-180. Berlin: De Gruyter.
8
Words, concepts, referents, and perspectives: Aspects of
metaphor in discourse
Gerard Steen (VU University Amsterdam)
The cognitive-linguistic analysis of metaphor typically builds a two-dimensional picture
in which metaphor is a matter of language and thought. In this talk I will argue that we
need a more sophisticated approach in which distinctions are made between aspects of
metaphor that have to do with metaphor in language (‘words’), metaphor in thought
(‘concepts’ and ‘referents’) and metaphor in communication (‘perspectives’). I will
present illustrations of these distinctions and their importance with reference to data
from the VU Amsterdam Metaphor Corpus. The distinctions can be motivated by
generally psychological considerations of discourse processing. They can moreover be
subjected to a form of explicit and formally constrained semiotic analysis that is entirely
compatible with the cognitive-linguistic agenda. It is the aim of my talk to demonstrate
this approach to metaphor in some detail and to argue for its adoption in any usagebased data analysis of metaphor in discourse.
9
Theme Session Papers: Cognitive Linguistics and
Phenomenology
Organised by Ad Foolen (Radboud University Nijmegen) and Paul Sambre
(University of Leuven)
Discussant: Frank Brisard (University of Antwerp)
11
Intentionality. Phenomenological and cognitive aspects
Els Elffers (University of Amsterdam)
Intentionality is a key concept in Searle’s speech act theory. Speech acts express a
psychological state of the speaker (e.g. assertions express a belief, directives express a
desire) and this psychological state is intentional: it is about something (Searle 1983).
Many linguists incorporated Searle’s views into their own theory, among them cognitive
linguists (cf. Dirven and Verspoor (1999), Nuyts (2008)).
Through the concept ‘intentionality’, there is a relation between speech act theory
and phenomenology as developed by Brentano and Husserl around 1900. Brentano
borrowed the concept from mediaeval scholastics and developed it into a central
element of his philosophical psychology. During this reintroduction, the concept
acquired a new dimension in addition to aboutness, namely directedness (cf. Van Baaren
1996).
This development caused a heterogeneous application of the concept, depending on a
focus on aboutness or on directedness. Heterogeneity increased by widely divergent
answers to intentionality-related philosophical questions, e.g. about the alleged
teleological character of intentionality or about the validity of the phenomenological
method in analyzing intentional acts.
In my talk the focus will be on the role of phenomenological thought about
intentionality in the important “depsychologization” of linguistics during the first
decades of the 20th century. Due to insights into the intentional character of language
use, the earlier view of linguistic utterances as mere expressions of the speaker’s
successive mental occurrences, was abandoned in favor of the view that utterances are
essentially hearer-directed acts related to objective contents. A clear signal of this major
transition is Bühler’s requirement that definitions of the notion ‘sentence’ no longer
refer to mental occurrences preceding the utterance (cf. Paul’s and Wundt’s “genetische
Definitionen”) but to their effect, the appeal made to the listener
(“Leistungdefinitionen”) (cf. Bühler 1918: 16).
Ironically, Searle’s work is a step backwards in this respect. Although speech acts are
defined in terms of their illocutionary effects, they are at the same time regarded as
expressions of the speaker’s mental states. Intentionality no longer triggers an “appeal”
view of language use; the concept resurfaces only as a feature of the speaker’s psychical
situation, which is expressed in his utterances. Significantly, Searle only speaks about
13
“intentional states” not about “intentional acts”, which used to be the customary
expression.
How do cognitive linguists cope with this aspect of Searle’s theory? Answers seem to
be variable. One example: Nuyts (2008) reinterprets Searle’s ideas in a less
“psychologistic” way, but goes not far enough to attain a fully-fledged act-oriented view
of linguistic utterances.
References
Baaren, Robbert van (1996) ‘Brentano's intentionaliteitsbegrip en het ontstaan van de
psychologie als wetenschap’. ANTW 87, 136-147.
Dirven, René and Verspoor, Marjolijn (eds.) (1999) Cognitieve inleiding tot taal en taalwetenschap.
Leuven etc.: Acco.
Bühler, Karl (1918) ‘Kritische Musterung der neuern Theorien des Satzes’.Indogermanisches
Jahrbuch 6,1-20.
Nuyts, Jan (2008) ‘Qualificational meanings, illocutionary signals, and the cognitive planning of
language use’. Annual Review of Cognitive Linguistics 6, 185-207.
Searle, John (1983) Intentionality. An essay in the philosophy of mind. Cambridge etc.: CUP.
14
A linguistic handshake with phenomenology
Ad Foolen (Radboud University Nijmegen)
Roman Jakobson’s strong interest in phenomenology (see Daalder 1999) was triggered
by his search for a theoretical foundation of the distinction between phonetics and
phonology. Following de Saussure’s (2016) general dichotomy between substance and
forme, an analogical pairing should be available on the level of meaning. In other words,
if phonology:phonetics = semantics:x, we need to fill in x. In Bloomfield’s view, we can
only do this on the basis of a scientific description of the world. As long as this
description is only partially available, “the statement of meanings” will remain “the
weak point in language study” (1933: 140).
The present paper aims to show that phenomenology is the proper antidote against
Bloomfield’s pessimism, more specifically, the version of phenomenology that stresses
the central role of the human body in perception and conception, as proposed by
Merleau-Ponty (cf. Sambre 2012, Krawczak 2013). It is the phenomenological world as
we know it through our body, senses, and action, which is the substance in relation to
which linguistically coded meanings should be analyzed. This view was already taken by
Dutch linguists like Witte (1949) and Wils (1965), long before modern theories of
embodied cognition came to the same conclusion.
Blomberg (2014)’s study on ‘motion’ shows that a systematic distinction between
motion as experienced in the ‘life-world’ and motion as coded in different languages is a
productive point of departure for the semantic analysis of motion expressions. In the
present paper, I will follow a similar approach, focusing on the analysis of Dutch and
German idiomatic expressions with hand, like De handen uit de mouwen steken ‘to stick
your hands out of the sleeves’, German die Ärmeln hochkrämpeln, to give only one
example of many: the hand is the most productive body part for creating idiomatic
expressions. Phenomenological studies on the hand, like Depraz (2013) and Gallagher
(2013), and other papers in Radman (ed.), are taken as a point of departure to fill in the x
in the ‘semantics:x’ formula.
References
Blomberg, Johan. 2014. Motion in language and experience. Actual and non-actual motion in Swedish,
French and Thai. PhD, Lund.
Bloomfield, Leonard. 1933. Language. New York: Holt, Rinehart and Winston.
Daalder, Saskia. 1999. H.J. Pos (1898-1955): Studies over zijn filosofie van taal en taalwetenschap. PhD,
Amsterdam.
15
Depraz, Natalie. 2013. Phenomenology of the hand. In: Z. Radman (ed.), 185-206.
Gallagher, Shaun. 2013. The enactive hand. In: Z. Radman (ed.), 209-225.
Krawczak, Karolina. 2013. Epiphenomenal semantics: Cognition, context and convention. Frankfurt
a.M.: Peter Lang.
Radman, Zdravko ed. 2013. The hand, an organ of the mind. What the manual tells the mental.
Cambridge, Mass.: The MIT Press.
Sambre, Paul. 2012. Fleshing out language and intersubjectivity: An exploration of MerleauPonty’s legacy to cognitive linguistics. Journal of Cognitive Semiotics 4(1), 189-204.
Saussure, Ferdinand de. 1916. Cours de linguistique générale. Paris: Payot.
Wils, J. 1965. Merleau Ponty’s taalbegrip en de moderne linguïstiek. Algemeen Nederlands
Tijdschrift voor Wijsbegeerte en Psychologie 57(2), 1-32.
Witte, J.J. de. 1948. De betekeniswereld van het lichaam. Taalpsychologische, taalvergelijkende studie.
[The meaning world of the body. Language psychological, language comparative
study]. PhD, Nijmegen.
16
Empathetic ambiguity and phenomenological immediacy
Dirk Geeraerts (University of Leuven)
As pointed out as early as Geeraerts (1985, 1993), a major point of appeal of the
phenomenological tradition for cognitive linguistics is its suggestion of an
epistemological immediacy that avoids the pitfalls of the Cartesian cogito. With the
growing attention in cognitive linguistics for the social nature of language, this overall
attractiveness may be further focused on the cultural and intersubjective aspects of the
phenomenological tradition: see e.g. Sambre (2012) on the cultural orientation of
Merleau‐Ponty's conception of phenomenological embodiment. Continuing that line,
the present talk will focus on the phenomenological conception of empathy, which in
recent years has been revived under the influence of mirror‐neuron research (see
Stueber 2006), and which ‐ in contrast with a.o. an analogy‐based theory of empathy ‐
suggests an unmediated, transparent presence of the Other.
However, the present talk will argue that the systematic ambiguities triggered by
second order empathy (a pervasive but understudied phenomenon) are a warning
against too much optimism about the intersubjective immediacy of empathy. If
empathy is the ability of Self to take into account Other’s point of view, then the ability
of Self to take into account Other’s point of view as including a view of Self (and
specifically, as including an empathic view of Self) may be characterized as 'second
order empathy'. Second order empathy exists in different mental modes (visual,
cognitive, verbal), but specifically, verbal second order empathy leads to systematic
ambiguities of interpretation, at diverse levels of linguistic meaning. The paper, then,
will explore the consequences of these ambiguities for a phenomenological
interpretation of intersubjectivity.
References
Geeraerts, Dirk. 1985. Paradigm and Paradox. Explorations into a Paradigmatic Theory of Meaning and
its Epistemological Background. Leuven: Universitaire Pers.
Geeraerts, Dirk. 1993. Cognitive linguistics and the history of philosophical epistemology. In
Brygida Rudzka‐Ostyn and Richard Geiger (eds.), Conceptualizations and Mental Processing
in Language 53‐79. Berlin/New York: Mouton de Gruyter.
Sambre, Paul. 2012. Fleshing out embodied language and intersubjectivity: an exploration of
Merleau‐Ponty's legacy to cognitive linguistics. Cognitive Semiotics 4: 189‐224.
Stueber, Karsten R. 2006. Rediscovering Empathy. Agency, Folk Psychology, and the Human Sciences.
Cambridge, Mass.: The MIT Press.
17
Interaction and intersubjectivity: the phenomenology of
multimodal co-constructions in instrumentals Submitted for
panel on cognitive linguistics and phenomenology
Paul Sambre (University of Leuven)
General objective - In this paper we try to relate some fundamental
phenomenological considerations, older and more recent, about intersubjectivity and
the perceptive status of material objects, to recent cognitive linguistic work in
multimodal construction grammar (MM CxG).
Framework - This work in construction grammar claims the multimodal nature of
constructions: constructions contain a verbal layer on top of which other, paraverbal or
non-verbal (visual) layers, like prosody, gesture, facial expressions and enacted object
manipulation may contribute to building rich conceptual structures. Furthermore, these
multimodal constructions may be distributed over different conceptualizers involved in
jointly speaking about a shared phenomenal world.
Data - We will elaborate on the conceptual status of words and objects by means of
which otherwise private conceptualizations acquire intersubjective status. As human
agents manipulate or conceive of objects, like instruments used for cutting and breaking
other objects, the body, through its posture and gestures, occupies a pervasive
conceptual position. In this rich phenomenological and linguistic view, real-life usage
events add experience to the pre-existent meanings of language, when speakers,, for
instance, creatively select argument structures for instrumental roles (DuBois 2003) or
exploit different conceptual entities for instrumentals, in varied constructions used in
perceiving and manipulating these objects, but also through the selection of
construction types, which interacts with how these speakers anticipate and respond to
their interlocutor involved in the action and talk about (represented) action (Lerner
2002, Ford & Thompson 1996).
Method - The phenomenological tradition in philosophy, where these notions of
body, interaction and intersubjectivity have been thorougly discussed, by MerleauPonty (1969), Alfred Schutz (2011) or, more recently, interaction theory (Gallagher &
Zahavi 2008), serves both as an input and endpoint to a reflection on multimodal
grammatical co-constructions. This phenomenological tradition, though, has been
largely ignored by cognitive linguistics.
Expected result - If grammatical descriptions, for instance, empirically reflect the
intersubjective multimodal event in which constructions are set up, and if the material,
corporeal and dialogic setting does affect the selection of constructions used for
building co-conceptualizations of usage events through language, then our linguistic
18
conceptualization of the world cannot be detached from a phenomenological account of
bodies in interaction.
References
Du Bois, J. 2003. Argument structure. Grammar in use. In Preferred Argument Structure. Grammar
as architecture for function, J. DuBois, L.E. Kumpf, W. J. Ashby (eds.), 11-60. Amsterdam:
Benjamins.
Ford, C., Thompson, S. 1996. Interactional units in conversation: syntactic, intonational and
pragmatic resources for the management of turns. In Interaction and grammar, E. Ochs,
E. Schegloff, S. Thompson (eds.), 134-184. Cambridge: CUP.
Gallagher, S., Zahavi, D. 2008. The Phenomenological Mind: An Introduction to Philosophy of Mind and
Cognitive Science. Abingdon: Routledge.
Lerner, G. 2002. Turn-Sharing: The Choral Co-Production of Talk-in-Interaction. In The Language
of Turn and Sequence, C. Ford, B. Fox, S. Thompson (eds.), 225-256. Oxford: OUP.
Merleau-Ponty, M. 1969 Signes. Paris: Gallimard.
Rodriguez, C., Moro, C. Coming to agreement: Object use by infants and adults. In The Shared
Mind. Perspectives on Intersubjectivity, J. Zlatev, T. Racine, C. Sinha, E. Itkonen (eds.), 89114. Amsterdam: Philadelphia.
Schutz, A. 2011. Collected Papers V. Phenomenology and the Social Sciences. Ed. by L. Embree.
Heidelber: Springer.
19
Regular Session Papers
21
Flies Flying around Lollipops: Metaphorical Conceptualization
of Islamic Hijab in Iran
Mohsen Bakhtiar (Eötvös Loránd University Budapest)
The proper observance of Islamic hijab is a serious issue for political-religious
hardliners in Iran. The concern revolves around the extramarital sexual attractionsexual arousal bond as the basis of individual and social corruption as well as the threat
of cultural metamorphosis through following a western lifestyle. In order to deal with
the problem, the religious fanatics resort to metaphorical language, among other
measures, to persuade Iranian women to modestly dress in public. The present paper,
based on Kövecses’s (unpublished manuscript) theory of metaphor in context, explores
the role of contextual factors involved in the formation of hijab linguistic metaphors
used in pro-hijab billboards and similar hijab-encouraging expressions. The frame
analysis of HIJAB revealed a tripartite classification based on which the hijab status of
women is evaluated. Moreover, analyzing hijab metaphors indicated that the
metaphorical conceptualization of women as FLOWER, PEARL, and BIRD act more
successfully in persuading a larger number of women to veil. On the other hand,
WOMEN AS CANDIES, or WOMEN AS SOLDIERS attracts a smaller category. Finally, the
metaphorical patterns represent the context-dependency of metaphorical meaning,
especially, the contextual role of ideology and entrenched conventional conceptual
metaphors in the production of novel metaphors as Kövecses’s theory suggests.
References
Cameron, L. (2003). Metaphor in educational discourse. London: Continuum.
Cameron, L. (2007). Patterns of metaphor use in reconciliation talk. Discourse and Society, 18,
197-222.
Cameron, L., & Deignan, A. (2006). The emergence of metaphor in discourse. Applied Linguistics.
27(4), 671-690.
Deignan, A. (2005). Metaphor and corpus linguistics. Amsterdam: John Benjamins.
Dirven, R, et al. (2007). Cognitive linguistics, ideology, and critical discourse analysis. In D.
Geeraerts, & H. Cuyckens (Eds.). The Oxford handbook of cogntive linguistics. (pp.
1222-1240). Oxford: Oxford University Press.
Gibbs, R. W., & Cameron, L. (2008). The social-cognitive dynamics of metaphor performance.
Journal of Pragmatics, 9(1-2), 64-75.
Kövecses, Z. (2002). Metaphor: A practical introduction. Oxford and New York: Oxford
University Press.
Kövecses, Z. (2006). Language, mind, and culture. A practical introduction. New York: Oxford
University Press.
Kövecses, Z. (2012). Creating metaphor in context. In S. Kleinke, Z. Kövecses, A. Musolff, and V,
Szelid (Eds). Cognition and culture. (pp. 28-43). Budapest: Szechenyiplan.
Kövecses, Z. Unpublished manuscript. Metaphorical Mind in Context.
23
Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chicago: The University of Chicago Press.
Lakoff, G. (1987). Women, fire, and dangerous things. Chicago: The University of Chicago Press.
Lakoff, G. (1996). Moral politics. What conservatives know that liberals don’t. Chicago:
University of Chicago Press.
Lakoff, G., & Johnson, M. (1999). Philosophy in the flesh. New York: Basic Books.
Musolff, A. (2006). Metaphor scenarios in public discourse. Metaphor and Symbol, 21(1): 23-38.
Van Dijk, T. A. (1991). Racism and the Press. London: Routledge.
Van Dijk, T. A. (1995). Discourse analysis as ideology analysis. Language and Peace. 17, 17-33.
Van Dijk, T. A. (2008). Discourse and context: A socio-cognitive approach. Cambridge: Cambridge
University Press.
24
Modeling Language Change with Construction Grammar
Jóhanna Barðdal (Ghent University)
In contrast to typologically well-known causative–anticausative pairs where a
derived anticausative has a nominative subject, Icelandic exhibits anticausatives where
the subject unexpectedly maintains the object case of the causative (with the direction
of the derivation being evident from the case marking of the arguments).
(1a)
Búnaðurinn …
skaut bátnum
upp eftir að skipið
sökk.
Equipment.the.NOM shot boat.the.DAT up after at ship.the sank
‘The equipment … shot the boat back up after the ship sank.’
Nom-Dat
(1b)
Öllum skaut upp úr ólgunni fyrir rest.
all.DAT shot up from swell for
rest
’Everybody came to the surface out of the swell in the end.’
Dat
On traditional and generative accounts of grammar, such alternations would be
conceptualized as a rule deriving one from the other synchronically (Zaenen & Maling
1990). On a constructional approach, however, the two would be regarded as separate
argument structure constructions standing in a systematic relation with each other,
with the intransitive denoting an event that has occurred spontaneously with no causer,
while the transitive is a causative with an explicit agent (Haspelmath 1997, inter alia),
shown in Figure 1. (based on Sandal 2011, who adopts the formalism of Goldberg 1995).
The verb skjóta means ‘emerge to the surface’ instead of ‘shoot up’. All these properties
must be specified as inherent features of this constructional alternation.
Alternations of this type are not confined to Modern Icelandic, but have been
documented in several early/archaic Indo-European languages such as Old NorseIcelandic (Westvik 1994, Sandal 2011, Ottósson 2013), Latin (Matasović 2013), Baltic and
Slavic. An ongoing change is found in Icelandic where the object case marking of the
anticausative subject becomes nominative; alternations of this type also disappeared in
Latin, long before the breakdown of the case system.
25
The two constructions are connected with an inheritance link, here labeled Ia for
“Inheritance” and “Anticausative”, making anticausative inheritance links on par with
other types of links that Goldberg posits, like instance/taxonomy, subpart, polysemy
and metaphorical links. Since constructions have multiple inheritance, the
Anticausative is also an instance of the ordinary intransitive construction, which means
that it inherits properties from that construction as well as from the transitive
construction. The intransitive construction, in turn, consists of several
subconstructions, namely Nom-only, Acc-only, Dat-only and Gen-only (Barðdal 2009).
Figure 1 above models an earlier stage of the Germanic languages, although the
alternation between the causative and this anticausative is not productive anymore,
representing a development from a systematic alternation to a stage where no link
between the two constructions is found. Instead, one is an instance of the transitive
construction and the other of the intransitive construction. The change from ProtoGermanic to Modern Icelandic can thus easily be modeled through a simple
disappearance of a link from the grammar.
26
References
Barðdal, Jóhanna. 2009. The Development of Case in Germanic. In J. Barðdal & S. L. Chelliah
(eds.) The Role of Semantic, Pragmatic, and Discourse Factors in the Development of Case, 123–
159. Amsterdam: John Benjamins.
Goldberg, Adele E. 1995. Constructions: A Construction Grammar Approach to Argument Structure.
Chicago: University of Chicago Press.
Haspelmath, Martin. 1987. Transitivity Alternations of the Anticausative Type. Universität zu Köln
(Arbeitspapier Nr. 5 (Neue Folge)).
Matasović, Ranko. 2013. Latin paenitet me, miseret me, pudet me and active clause alignment in PIE.
Indogermanische Forschungen 118: 93–110.
Ottósson, Kjartan. 2013. The Anticausative and Related Categories in the Old Germanic
Languages. In Folke Josephson & Ingmar Söhrman (eds.) Diachronic and Typological
Perspectives on Verbs, 329–381. Amsterdam: Benjamins.
Sandal, Catrine. 2011. Akkusative subjekt og antikausativitet i norrønt [Accusative Subjects and
Anticausativity in Old Norse]. University of Bergen M.A. Thesis.
Westvik, Olaf Jansen. 1994. On the subjects of some nominativeless sentences in Old Germanic.
In T. Swan, Endre Morck, & Olaf Westvik (eds.) Language Change and Language Structure:
Older Germanic Languages in a Compar-ative Perspective, 305–344. Berlin: De Gruyter.
Zaenen, Annie & Joan Maling. 1990. Unaccusative, Passive and Quirky Case. In A. Zaenen & J.
Maling (eds.) Modern Icelandic Syntax, 137-152. Academic Pres.
27
A usage-based model of early grammatical development
Barend Beekhuizen (Leiden University)
A striking aspect of language acquisition is the difference between children's and
adult's utterances. Not only are many functional elements (determiners, auxiliaries)
absent from early productions, we also encounter verb-valency patterns more restricted
in scope than typically encountered with adults.
Although much research in the usage-based tradition has been devoted to the limited
productivity of early grammatical representations (see Ambridge & Lieven 2012: ch. 6
for an overview; we could call this the limitedness of the paradigmatic dimension), the
representations and processes yielding the limited verb-valency patterns of early
language production have received less attention (which we could call the syntagmatic
dimension). This research aims to fill that theoretical gap, using a computational
operation of the learning processes.
Building on earlier proposals (Schlesinger 1971, Braine 1976), and inspired by
Langacker’s (2009) exposition of language-acquisition-as- processing, we propose a
computational model that acquires form-meaning pairings from noisy input data. We
evaluate the model’s performance both on its ability to understand novel sentences,
and, crucially, to produce utterances on the basis of unseen semantic representations.
We show that both the increasing length and productivity found in children’s
productions can be simulated.
To the best of our knowledge, this is the first computational model that both involves
symbolic pairings of phonological and semantic material, and is able to simulate the
gradual development of children’s grammatical productions. Moreover, whereas many
other models of grammatical acquisition assume lexical knowledge to be in place, our
model starts with an empty set of linguistic representions. On the linguistic-theoretical
side, the proposed mechanisms form a further specification and formal
operationalization of Langacker’s (2009) view on language acquisition and fill a blind
spot of much usage-based theorizing, namely the syntagmatic development.
References
Ambridge, Ben & Elena V.M. Lieven (2012). Child Language Acquisition. Contrasting Theoretical
Approaches. Cambridge, Cambridge University Press.
Braine, Martin D.S. (1976). Children’s first word combinations. Monographs of the Society for
Research in Child Development, 41(164): 1-104.
Langacker, Ronald W. (2009). A Dynamic View of Usage and Language Acquisition. Cognitive
Linguistics, 20: 627-640.
28
Schlesinger, I.M. (1971). Production of Utterances and Language Acquisition. In Dan I. Slobin
(ed.) The Ontogenesis of Grammar. A Theoretical Symposium, p. 63-102. New York, Academic
Press
29
Two types of Dutch intransitive verbs? The role of attested
data
Maaike Beliën (Delft University of Technology)
In line with the Unaccusative Hypothesis (Perlmutter 1987, Burzio 1984), Dutch is
often assumed to have two types of intransitive verbs: unaccusatives, such as vallen ‘fall’,
and unergatives, such as werken ‘work’. The two types of verbs are considered to show
different syntactic behavior, as illustrated by examples (1)-(3) from Alexiadou et al.
(2004: 9). Such data support the idea that past participles of unaccusative verbs cannot
occur in impersonal passives, they take zijn as their auxiliary in the perfect tense, and
they cannot be used as adjectives before a noun. Past participles of unergative verbs, on
the other hand, can occur in impersonal passives, they take hebben as their perfect
auxiliary, and they can be used as attributive adjectives.
(1) Er wordt *gevallen/gewerkt
there is fallen/worked
(2) De jongen is gevallen/*gewerkt
the boy is fallen/worked
(3) de gevallen/*gewerkte jongen
the fallen/worked boy
A number of recent studies, however, have presented attested examples, mainly
collected from the Internet, that are problematic for these generalizations. Primus
(2011) and van Schaik-Rădulescu (2011), for example, provide examples of Dutch
impersonal passives with verbs that have been classified as unaccusatives, cf. e.g. (4)
from van Schaik-Rădulescu (2011: 71). Beliën (2012) discusses examples of auxiliary
choice with Dutch manner of motion verbs that cross-cut the unaccusative-unergative
distinction. Elffers et al. (2012), finally, present empirical data to support their intuition
that some unaccusative verbs are quite restricted in their use as attributive adjectives,
such as #de gestopte man ‘lit. the stopped man’.
(4) Rond km.38 zat ik iets meer achterin het peloton toen er vlak voor mijn neus gevallen
werd.
‘At about km 38 I was riding a bit farther in the back of the pack when there
was a fall right in front of my nose.’
These studies appear to have developed largely independently from each other, each
focusing on only one of the grammatical constructions deemed relevant to the
30
Unaccusative Hypothesis. With their use of attested data, however, the studies
represent a methodological departure from earlier work in this area, which relied on
constructed examples with grammaticality judgments (cf. (1)-(3) above). On the basis of
a review of the recent work, this presentation aims to identify the implications of this
methodological departure for studying linguistic phenomena related to the
Unaccusative Hypothesis.
31
Collexemes, constructions, and polysemy: zullen (‘will’/‘shall’)
and gaan (‘be going to’)
Ronny Boogaart & Robbert van Etten (Leiden University)
Right from the start of cognitive linguistics, polysemy has been a major concern, and
at the same time a major claim, of studies dealing with modal verbs. As such, cognitive
linguistic accounts of modals have always focused on the conceptual-semantic side of
modality. The more recent advance of construction grammar may be seen as
complementary to such purely semantic accounts. In this framework, different
functions of a modal verb are ascribed to systematic differences in the linguistic context
in which it is used. These larger patterns, or modal constructions, act as constraints on
the polysemy of the modal verb and are themselves not necessarily polysemous at all.
But how can we determine all the constructions that a given modal verb is a
constituting part of, and what exactly does it tell us about the meaning of the verb?
In this talk, we investigate to what extent the method of collostructional analysis is
capable of answering such questions. The case in point is constituted by the Dutch verbs
zullen (‘will’/‘shall’) and gaan (‘be going to’). A classic (recently revived) discussion
concerns the question whether zullen is an auxiliary of future time or a modal verb.
Since both uses are attested, it seems to be a school book example of polysemy, but the
Dutch linguistic tradition consistently argues in favour of a monosemous analysis in
which zullen is considered a modal verb expressing epistemic modality. As for gaan, it
has been argued that it is developing from a motion verb to a form expressing future
time and in the latter function it may compete with zullen. Therefore, the two verbs
make interesting candidates for a more specific distinctive collexeme analysis. We
present the results of such an analysis conducted in the written language Corpus
Hedendaags Nederlands (Corpus of Contemporary Dutch) of the Instituut voor
Nederlandse Lexicologie (INL), available on-line since 2014. It turns out that there is
hardly any overlap between the infinitives (collexemes) attracted to zullen and gaan. A
further interesting outcome is that the collexemes of zullen and gaan differ significantly
depending on tense and person/number of the finite form.
In the final part of our talk we will reflect on the relationship between the collexemes
that result from our corpus-based research, and the issues that are central to the more
theoretical discussions about modal verbs: What, if anything, does collostructional
analysis tell us about the relationship between the different readings of zullen and gaan,
and more generally about the alleged polysemy or monosemy of modal verbs?
32
The pluralization of presentational haber in Caribbean Spanish.
A study in Cognitive Construction Grammar and comparative
sociolinguistics
Jeroen Claes (University of Antwerp)
In standard Spanish, presentational haber (‘there-to-be’) is an impersonal
construction: it only takes the third person singular verb-ending and its nominal
argument, e.g. mangos ‘mangoes’ in (1), behaves as a direct object. However, there
seems to be a growing tendency to establish verb-agreement with the NP (cf. D'AquinoRuiz, 2008), which is known as 'the pluralization of haber'. From a construction-grammar
perspective (cf. Goldberg 1995, 2006), this could indicate that the singular construction
(<AdvP haber Object>) coexists with a pluralized schema (<AdvP haber Subject>). I
hypothesize that in Caribbean Spanish, this alternation constitutes an ongoing linguistic
change from below. Additionally, I present the hypothesis that haber pluralization is
constrained by three general cognitive factors: markedness of coding, statistical
preemption, and structural priming.
(1) Sí, había mangos (SJ14H22 /SJ1672).
‘Yes there wereSing mangoes.’
(2) En mi época habían unos veinticinco, treinta alumnos por aula
(LH01H22/LH17).
‘In my days, there werePlur some twenty, thirty students per classroom.’
To evaluate these hypotheses, I present a variationist analysis. In three recent
samples of the varieties of Havana, Santo Domingo, and San Juan, I trace the
grammatical and social distribution of the alternation, in order to establish whether and
to which extent the variation points to an ongoing linguistic change from below and an
alternation involving two variants of the presentational construction with haber that is
constrained by these three general cognitive factors.
Overall, the results indicate that speakers of Caribbean Spanish pluralize the verb in
similar proportions (Havana: 44.6%, N=934/2093; Santo Domingo: 46.7%, N=859/1841;
San Juan: 41.4%, N=682/1649). The comparative sociolinguistic analysis (involving
mixed-effect logistic regression, conditional inference trees, and random forests)
displays little variation when it comes to grammatical constraints: in the three capitals,
the pluralization is favored by NPs that refer to typical action-chain heads, the use of
the pluralized variant in prior discourse and all verb tenses, minus the present and
33
preterit. The sole difference is that, only in San Juan does the absence of negation
support pluralization.
By contrast, the associations between haber pluralization and social groups vary more
significantly according to the respective speech communities. In Havana, the
alternation is linked to lower social class. In Santo Domingo, haber pluralization
expresses middle-class membership. In San Juan, finally, the pluralization of haber is
associated to female gender. The results also suggest a significant interaction between
age and social class, in the sense that younger speakers appear to associate haber
pluralization to lower social class, whereas it is a typical feature of middle-class speech
among older speakers. For none of the three communities, the data display styleshifting.
These results suggest that the pluralization of haber constitutes an advanced change
from below (Labov, 2001: 308-309), which is constrained by the three general cognitive
factors.
References
D'aquino-Ruiz, Giovana. (2008). El cambio linguistico de haber impersonal. Núcleo, XX (25), 103124.
Goldberg, Adele. (2006). Constructions at work. The nature of generalization in language. Oxford:
Oxford University Press.
Goldberg, Adele. (1995). Constructions. A construction-grammar approach to argument-structure.
Chicago: Chicago University Press.
Labov, William. (2001). Principles of linguistic change. Volume 2: social factors. Oxford: Blackwell.
34
A reassessment of the diachrony of the auxiliary selection
alternance in Dutch verbal clusters with a modal
Evie Coussé (University of Gothenburg) & Freek Van de Velde (University of Leuven)
Dutch features a conspicuous alternance in its auxiliary selection in clusters with a
modal verb: both (1) and (2) occur.
(1)
Hoe
heeft
dat
kunnen
how
has
that
can
‘How could this have happened’
gebeuren?
happen
(2)
Hoe
is
dat
kunnen
how
is
that
can
‘How could this have happened’
gebeuren?
happen
The reason behind this variation is that in variant (1), the perfective auxiliary is
selected by the modal (kunnen ‘can’), which in Dutch invariably takes hebben (‘have’) as
its auxiliary when it occurs in isolation, see (3), whereas in variant (2), it is the main
verb gebeuren (‘happen’) that selects the perfective auxiliary. Gebeuren is an
unaccusative that selects zijn for its perfect tense in isolation, see (4).
(3)
Hij
heeft
het
he
has
it
‘He was able to do it’
(4)
Het
is
it
is
‘It happened’
gekund
can:PTCP
gebeurd
happened
In the literature, it has been claimed that (2) is an innovative variant, slowly gaining
ground on (1) (Van der Horst 1998, Van der Horst & Van der Horst 1999, Draye & Van
der Horst 2006, Van der Horst 2008). Remarkably, this would mean that the South,
where variant (1) is predominant, is ‘ahead’ of the North, where variant (2) is preferred.
In our talk, we want to critically look at the diachronic evidence supporting this claim.
On the basis of existing observations, and of additional diachronic corpus research
(using the corpus of Coussé 2010), we argue that the evidence for the scenario described
above is not very solid. We propose an alternative, construction-based account, that fits
better with the diachronic data. In essence, this account entails that both variants are
35
the result of the extension and integration of different source constructions, but each of
the variants resolves the conflicts arising from the integration in a different way.
References
Coussé, Evie. 2010. Een digitaal compilatiecorpus historisch Nederlands. Lexikos 20. 123-142.
Draye, Luk & Joop van der Horst. 2006. Hebben/haben versus zijn/sein in drieledige
werkwoordsgroepen in het Duits en het Nederlands. In Matthias Hüning, Arie
Verhagen, Ulrike Vogl & Ton van der Wouden (reds.), Nederlands tussen Duits en Engels,
229-242. Leiden: Stichting Neerlandistiek Leiden.
Hofmans, Mark. 1982. Hebben of zijn: of de deverbalisering van de modale werkwoorden in het
Nederlands. Linguistics in Belgium 5. 81-109..
Van der Horst, Joop. 1998. Over de geschiedenis van de Nederlandse hulpwerkwoorden. In Wim
Klooster, Hans Broekhuis, Els Elffers-van Ketel & Jan Stroop (reds.), Eerste Amsterdams
colloquium Nederlandse taalkunde. Lezingen gehouden op 10 en 11 december 1997, 101-119.
Amsterdam: Nederlandse Taalkunde UvA.
Van der Horst, Joop. 2008. Geschiedenis van de Nederlandse Syntaxis. Leuven: Universitaire Pers
Leuven.
Van der Horst, Joop & Kees van der Horst. 1999. Geschiedenis van het Nederlands in de twintigste
eeuw. ’s-Gravenhage: SDU.
36
Gender-neutral or not: feminine forms as self- reference in
Spanish
Barbara De Cock (Université catholique de Louvain)
The Spanish personal pronoun paradigm allows for a typologically rather rare
(Siewierska 2004) gender distinction in the 1st person plural, viz. plural nosotras ‘we’
refers to an all-female group (instead of nosotros). In addition, the indefinite pronoun
uno ('one'), which is often associated with speaker-reference, also has a feminine
counterpart una. Whereas the grammatical tradition is quite unanimous as to the use of
nosotras by women, the use of una, however, has been described in a more nuanced way,
with some authors highlighting regional differences (Butt & Benjamin 1988, NGLE 2010)
or differences related to the degree of self-reference (Butt & Benjamin 1988) or
expression of the own opinion (NGLE 2010:1132). I will argue that the use of feminine
forms for self-reference is not only related to the sex of the speaker and to selfreference, but also to the conceptualization she wishes to convey, relying crucially on
the concepts of subjectivity and intersubjectivity (Benveniste 1966, Lyons 1994, Traugott
2003)
In this paper, I will analyze the use of both nosotras and una in contexts where a
female identity is particularly at stake, namely conversations, debates and blogs on
women's rights or other topics that are typically associated with women, such as
pregnancy. From this analysis, it will become clear that women may refrain from using
the feminine forms in view of certain conceptualization choices, namely when their
female identity seems less at stake or, crucially, to achieve an intersubjective effect
when affiliation with male interlocutors is being sought.
In the second place, I will look into the differences between nosotras and una, as it
seems that the more direct speaker inclusion realized by 1st person plural pronoun
nosotras makes it less likely for women to refrain from using the feminine form. The
vaguer speaker inclusion realized by indefinite uno/una, on the contrary, seems to
license more easily the use of masculine uno, even for reference to a female speaker or
group.
Thus, I will show that the use of feminine pronouns is not only determined to the sex
of the speaker, but that female speakers may conceptualize different subjective
positions and create intersubjective effects through pronoun choice.
37
References
Benveniste, Emile. 1966. “De la subjectivité dans le langage.” In Problèmes de linguistique Générale,
ed. by Emile Benveniste, 258-266. Paris: Gallimard.
Butt, John, & Carmen Benjamin. 1988. A new reference grammar of modern Spanish. London: Arnold.
NGLE: Real Academia Española & Asociación de Academias de la Lengua Española. 2010.
Nueva Gramática de la Lengua Española.Madrid: Espasa.
Lyons, John. 1994. “Subjecthood and Subjectivity.” In Subjecthood and Subjectivity. The Status of the
Subject in Linguistic Theory, ed. by Marina Yaguello, 9-17. Paris: Ophrys.
Siewierska, Anna. 2004. Person. Cambridge Textbooks in Linguistics. Cambridge: Cambridge
University Press.
Traugott, Elisabeth C. 2003. “From Subjectification to Intersubjectification.” In Motives for
Language Change, ed. by Raymond Hickey, 124-39. Cambridge: Cambridge University
Press.
38
Parallel sources, different outcomes: a corpus-based study of
the ‘far from X’ construction in Dutch, English and French
Hendrik De Smet (University of Leuven), Kristel Van Goethem (Université catholique
de Louvain) & Gudrun Vanderbauwhede (Université de Mons)
In language change, parallel source constructions can undergo cross-linguistically
divergent devel-opments. The focus of this paper is on one such case, the development
of degree modifiers from markers of physical distance. Specifically, we compare the
histories of Dutch verre van, English far from and French loin de – all three consisting of an
adjective/adverb and a preposition, and all three meaning ‘far from’.
The first purpose of our study is to analyze to what extent the parallel source
constructions have developed degree modifying uses. Data from the BNC show that
English far from can be used as de-gree modifier with gerunds and adjectives (cf. De Smet
2012).
(1) this competition, far from resolving the problem of security, in fact
exacerbates it (BNC)
(2) Nutty was far from sure, and Biddy looked doubtful. (BNC)
Data from the COW corpus (Schäfer & Bildhauer 2012) show that Dutch verre van has
undergone constructional specialization: it acts as an adverbial downtoner in more than
87% of the cases (3), whereas its formal variant ver van is exclusively used to indicate
spatial or metaphorical distance (4). Moreover, contrary to English far from, verre van can
occur without a complement (5), signaling an even more advanced degree of
adverbialization.
(3) Nee, dit is verre van plezierig. (NLCOW 2012) ‘No, this is far from pleasant.’
(4) Iquitos ligt niet zo ver van Brazilië en dat is te merken. (NLCOW 2012) ‘Iquitos
is not so far from Brazil, and that shows’.
(5) Ik zeg niet dat hier alles beter is, verre van, zou ik haast zeggen. (NLCOW 2012)
‘I don’t say that everything is better here, far from (it), I would almost say.’
Contrary to Dutch, French loin de combines the spatial, metaphorical (6) and degree
modifying use (7). However, syntactically loin de does not act as an adverb, but always
functions as a complex prep-osition (with nominal complements), even if used with a
downtoning function (7).
39
(6) Nous voilà loin de la mondialisation heureuse ! (FRCOW 2011) ‘Here we are far
from happy globalization!’
(7) Or, c’est loin d’être le cas. (FRCOW 2011) ‘But this is far from being the case.’
As a second objective, we aim to account for these synchronic differences by
exploring the dia-chronic pathways taken by [[far from] [X]]YP [[ver/verre van] [X]]YP and
[[loin de] [X]]PP and by examin-ing if their diachronic development into adverbial
/prepositional downtoners can be seen as an in-stance of (grammatical)
constructionalization (Traugott & Trousdale 2013). For this purpose, we will carry out
an in-depth diachronic corpus study, based on data that we have already compiled from
EEBOCorp for English, Frantext for French and the KB newspaper archive for Dutch.
These diachronic data will be interpreted in terms of the parameters proposed for
grammatical constructionalization by Traugott and Trousdale (2013), that is
productivity, schematicity and compositionality.
References
De Smet, H. (2012). The course of actualization. Language 88. 601-633.
Schäfer, R. & F. Bildhauer. (2012). Building large corpora from the web using a new efficient tool
chain. N. Calzolari, K. Choukri, T. Declerck et al. (Eds), Proceedings of the Eight
International Confer-ence on Language Resources and Evaluation, Istanbul, 486-493.
Traugott, E. C. & G. Trousdale (2013). Constructionalization and Constructional Changes. Oxford: Oxford University Press.
Corpora
BNC = British National Corpus: http://corpus2.byu.edu/bnc/
COW = Corpora from the Web: http://hpsg.fu-berlin.de/cow/colibri/
EEBOCorp: https://lirias.kuleuven.be/handle/123456789/416330
Frantext: http://www.frantext.fr/
KB: http://kranten.kb.nl/
40
Language is a bio-cultural hybrid: a cross-linguistic study of
English and Mandarin expressions of the Self from a cognitive
point of view
Simon Devylder (Jean Moulin University Lyon)
How do people structure abstractions? Time, for instance, is constructed through
analogical extensions of the more experience-based domain of space and its
conceptualization varies from one culture to another 11 . This paper will focus on an
equally ubiquitous abstraction we live through: the concept of Self. We collected two
sets of metaphorical expressions people frequently use to refer to themselves in English
and in Mandarin Chinese. On the one hand the set of English expressions show that its
speakers refer to the Self as a whole made of parts. For instance an entity whose unity
can be damaged (I’m breaking to pieces ; I’m falling apart) or preserved (I’m whole again ; A
man of integrity). On the other hand Mandarin speakers refer to the Self as an “inside”
and an “outside” that can be consistent (表里如一, Biao li ru yi: “outside inside like one” ;
秀外慧中, Xiu wai hui zhong: “beautiful outside wise inside”) or inconsistent (表里不一,
Biao li bu yi: “outside inside not one” ; 外 中干, wai qiang zhong gan: “outside hard inside
dry”). This cross-linguistic study reveals that when speakers of both groups express a
continuous relation (whether complete or consistent) between the constituents of
Selfhood (parts/whole ; inside/outside) they sense their Self as being harmonious,
whereas when the relation is discontinuous (whether incomplete or inconsistent) they
experience their Self as being chaotic. We extend this specific analysis to the wider
linguistic debate of language universals versus linguistic relativity, by arguing that
language is a bio- cultural hybrid2: an embodied symbolic system shaped by the human
biological machinery from which it emerges and also a vehicle for the worldview of a
given culture, constraining its speakers to select culture-specific linguistic strategies to
experience their Self. From this point of view we first propose the hypothesis that
English and Mandarin speakers both refer to the common primitive image schema
/SELF -CONTINUITY IS GOOD– SELF-DISCONTINUITY IS BAD/ because it was
Majid, A., Gaby, A. & Boroditsky, L. (2013). Time in terms of space. Frontiers in
Psychology, 4(554), online
2
Evans Nicholas and Levinson Stephen C., (2009). “The myth of language universals:
Language diversity and its importance for cognitive science”. Behavioral and Brain
Sciences 32: 429-492.
1
41
incorporated at a pre-linguistic level from an embodied experience undergone at an
early stage of childhood, where, to cope with the feeling of helplessness, an infant finds
in the mirror image a promise to be a consistent, pulled-together whole1. Secondly we
explain that it is well- identified cultural characteristics which encourage English
speakers to refer to themselves as a whole made of parts and Mandarin speakers to
comprehend Selfhood in terms of consistency.
1
42
Johnston Adrian, "Jacques Lacan", SEP (Summer 2013 Edition), Edward N. Zalta (ed.)
Deconstructing a verbal illusion: the ‘No X is too Y to Z’
construction and the rhetoric of negation
Egbert Fortuin (Leiden University)
Consider the following sentences which both have the structure NO X BE TOO Y TO Z:
(1) No head injury is too trivial to be ignored.
(2) No mountain is too high to climb.
At first glance we seem to be dealing with the very same construction, but closer
scrutiny reveals that the two sentences get a very, even diametrically opposed,
inference, namely that the situation expressed by the infinitive Z cannot or should not be
realized (1) or can or should be realized (2). It is usually assumed that sentences such as
(1) are incorrect (grammatical mistakes) but get a coherent interpretation because of
shallow processing, and because of pragmatic factors, which overrule semantics and
syntax (e.g. Wason and Reich 1979). In this paper I argue that this analysis is incorrect
and that (1) can be seen as an instance of the negative ‘No X is too Z to Y’ construction,
which is a sub-construction of the abstract ‘No X is too Y to Z’ construction (cf. the
notion of ‘family of constructions’ given in Goldberg and Jackendoff 2004). I will show
that in contrast to the negative sub-construction, the meaning of the positive subconstruction can be compositionally derived. Nevertheless, the negative construction
can be seen as conventionalized construction (form-meaning unit) that has a
transparent (i.e. linguistically analyzable) syntactic structure that can be linked to and
motivated by other constructions. As I will show, the occurrence of negative ‘No X is too
Z to Y’ construction has to do with the rhetorical function of the infinitival verb in these
sentences, and the need to express particular negative information by a form-meaning
element. As such, my analysis stresses the importance of the rhetorical dimension of
constructions (cf. Verhagen 2005).
References
Goldberg, Adele and Ray Jackendoff. 2004. The English resultative as a family of constructions.
Language 80(3), 532-568.
Verhagen, Arie. 2005. Constructions of intersubjectivity: discourse, syntax, and cognition. Oxford:
Oxford University Press.
Wason, P. and S.S. Reich. 1979. A verbal illusion. Quarterly Journal of Experimental Psychology
31:591-97.
43
Predicting the dative alternation in early modern Dutch
Tim Geleyn & Timothy Colleman (Ghent University)
Up until the late Middle Dutch period (1500 AD), the dative case was the standard way
in Dutch of marking the recipient argument in ditransitive scenes with various verbs of
giving and verbs of semantically related verb classes. Sentence (1) below provides an
example with the typical dative marker n.
(1) Mer men beval
hem den gilden. (Utrechtsch-Hollandsche jaarboeken, 1481-1483)
but they consign.PST him the guilds
“But they consigned him to the guilds.”
By the start of the early modern Dutch period (from 1500 onwards), the
dative/accusative distinction had already eroded (Van der Horst 2008: 578), and was
eventually lost. In the above mentioned ditransitive contexts, the locative preposition
aan (which is cognate with English on and German an) was adopted as recipient marker,
see e.g. (2) for an early 17th-century example.
(2) ... dat … het lichaem wederom sij aen den vrinden gaeven. (Achilles en Polyxena, 1600)
that the body back
they AAN the relatives give.PST
“... that they returned the body to the relatives.”
This prepositional construction was fully sanctioned by contemporary grammarians
in the 17th century (Weijnen & Gordijn 1970). From that period onwards, many
ditransitive verbs alternate between the double object construction (DOC, with zeromarked recipient argument) and the prepositional construction with aan – a
phenomenon commonly described as the ‘dative alternation’. The choice between these
two largely functionally equivalent argument structure constructions is nevertheless
not random, as has been shown in e.g. Van Belle & Van Langendonck (1995) and
Colleman (2009), or for the present-day English dative alternation in Bresnan (2007). In
this paper, we would like to determine to what extent the choice between both
constructions in the earlier periods was determined by the same (kinds of) factors
known to determine the dative alternation today. On the basis of a self-compiled corpus
of literary prose (1600-1650), we will perform a logistic regression to assess the strength
and predictive power of various explanatory variables, including semantic parameters
(e.g. semantic class of the verb, animacy of the arguments), formal parameters (e.g.
length of theme and recipient) and discourse parameters (e.g. givenness of theme and
recipient). In this way, we will try to provide new insight into the different factors at
44
work in verb pattern alternations, and reveal possible diachronic shifts in their relative
weight.
References
Bresnan, J., Cueni, A., Nikitina, T. & Baayen, R.H. 2007. Predicting the dative alternation. In
Boume, Kraemer & Zwarts (eds.). Cognitive foundations of interpretation. Amsterdam:
Royal Netherlands Academy of Science.
Colleman, T. 2009. Verb disposition in argument structure alternations: a corpus study of the
dative alternation in Dutch. Language Sciences, 31(5):593-611.
Van Belle, W., Van Langendonck, W. 1996. The indirect object in Dutch. In: Van Belle, W., Van
Langendonck, W. (Eds.), The Dative. Descriptive Studies, vol. 1. Amsterdam: John
Benjamins Publishing Company, 217–250.
Van der Horst, J. 2008. Geschiedenis van de Nederlandse Syntaxis. Leuven: Universitaire Pers
Leuven.
Weijnen, A. A. & Gordijn, A. M. J. 1970. Het binnendringen van het voorzetsel in het indirekt
object in het Nederlands. Leuvense Bijdragen, 59:125-130.
45
What remains? A constructivist examination of verbs in
aphasia
Rachel Hatchard & Ruth Herbert (University of Sheffield)
Verbs commonly present challenges for people with aphasia (Links et al. 2010). Much
research in this area has addressed verbs as single words (see Webster & Whitworth
2012) and studies examining verbs in wider structures are mostly underpinned by rulebased, generative theory (e.g. Bastiaanse & Thompson 2003). There appear to be no
constructivist characterisations of verbs in aphasia.
In the constructivist view, verbs are closely linked with the constructions they are
produced in through, for example, semantic association and frequency effects. In this
approach, language is acquired from the input, beginning with single-word and itembased constructions, before generalisations are made to form more schematic
abstractions of these items (Ambridge & Lieven 2011). Accordingly, verbs are acquired
in piecemeal fashion with each verb developing in its own time-scale rather than full
mastery of a verb being achieved outright (Tomasello 1992). Such characterisations in
acquisition raise questions about verbs in aphasia: which verbs remain and with what
level of abstraction? Also, how might these be influenced by frequency levels in the
input? Since more frequent items should be more entrenched, they may be easier to
retrieve and perhaps more resilient to loss. This would make them likely candidates for
retention and relatively easy to access by people with aphasia.
This study examines the verbs produced in Cinderella narratives by three adults with
aphasia, including one PATSy database participant (Lum et al. 2012). Specifically, it
investigates (1) the verb types (lemmas), (2) verb forms (tokens) and (3) the
constructional combinations in which verbs are produced. Particular attention is paid to
productivity levels and any link with frequency (in the Spoken British National Corpus,
Davies 2004-). Comparisons are also drawn with a group of twelve non-aphasic speakers
narrating the same story.
Preliminary results indicate that the verbs produced are indeed piecemeal and
mainly limited to a small number of concrete forms and constructional combinations.
Also, there are links with frequency at all levels (type, form, combination). These initial
findings support a constructivist approach and thus show how this could offer fresh
insight into verbs in aphasia.
46
References
Ambridge, B., & Lieven, E.V.M. (2011). Language acquisition: Contrasting theoretical approaches.
Cambridge: Cambridge University Press.
Bastiaanse, R., & Thompson, C.K. (2003). Verb and auxiliary movement in agrammatic Broca’s
aphasia. Brain and Language 84(2), 286-305.
Davies, M. (2004-). BYU-BNC (Based on the British National Corpus from Oxford University
Press). http://corpus.byu.edu/bnc/ [accessed April 2014].
Links, P., Hurkmans, J., & Bastiaanse, R. (2010). Training verb and sentence production in
agrammatic Broca’s aphasia. Aphasiology 24(11), 1303-1325.
Lum, C., Cox, R., & Kilgour, J. Universities of Sussex and Edinburgh. PATSy: A database of clinical
cases for teaching and research. http://www.patsy.ac.uk. [accessed January 2012].
Tomasello, M. (1992). First verbs: A case study of early grammatical development. Cambridge:
Cambridge University Press.
Webster, J. & Whitworth, A. (2012). Treating verbs in aphasia: Exploring the impact of therapy
at the single word and sentence levels. International Journal of Language & Communication
Disorders 47(6), 619–636.
47
Interactional shifts between the three terms of address in
everyday situations in Chilean Spanish
Kris Helincks (Ghent University)
This lecture focuses on the negotiation and variation of the three terms of address
(ToAs) in ongoing interaction in Chilean Spanish. The Chilean address system contains
three pronominal ToAs with corresponding verb forms: ustedeo (usted+3rd p.sg.), tuteo
(tú+2nd p.sg.) and voseo (vos+archaic 2nd p.pl.). Since the 1960, this threefold system
undergoes a linguistic change with an increasing use of verbal voseo opposed to a
persistent stigmatization of pronoun vos (Torrejón 1986). This evolution has recently
prompted some sociolinguistic studies which supply the first systematic empirical data
on familiar voseo vs. normative tuteo (Stevenson 2007, Rivadeneira 2009, Bishop &
Michnowicz 2010, Helincks 2012). As a natural consequence of this linguistic change in
progress, shifts between these ToAs are common, even within the same interaction.
Moreover, Chilean Spanish characterizes a ‘usted of tenderness’ with a specific
pragmatic use and meaning. Despite the common practice of interactional shifting
between the three ToAs in Chile (2010: 134-135), the above mentioned studies comment
only anecdotally on these shifts or explain them unsatisfactorily as ‘free’.
In this lecture I present quantitative results of a sociopragmatic-oriented research on
this topic. The data come from a corpus of spontaneous conversations in everyday
private and public situations. Social, relational and situational factors are analysed, as
well as the influence of normative use, facework and identity construction. The following
research questions will be answered: how frequent are shifts between the three ToAs
within an act, intervention, exchange or conversation between two interlocutors?;
which shifts are most common?; do some situations trigger more shifts than others?; do
age (difference), gender (difference) and type of relationship of interlocutors influence
the amount of shifting?; do certain shifts relate to specific facework strategies?; and, does
an individual address variation system express a particular social identity? The analysis
contributes to how contextual, pragmatic and discursive factors determine or favour
interactional negotiation and variation of ToAs.
References
Bishop, Kelley, Michnowicz, Jim (2010): “Forms of Address in Chilean Spanish”, Hispania93 (3):
413-429.
Helincks, Kris (2012): “La variación social y estilística del voseo chileno en diferentes géneros
televisivos”, Revista Internacional de Lingüística Iberorománica 19: 185-211.
48
Hummel, Martin (2010): “Reflexiones metodológicas y teóricas sobre el estudio de las formas de
tratamiento en el mundo hispanohablante, a partir de una investigación en Santiago de
chile”, in M. Hummel, B. Kluge, M.E. Vázquez Laslop, (eds.): Formas y fórmulas de
tratamiento en el mundo hispánico. México D.F./Graz: El Colegio de México/Karl Franzens
Universität: 101-162.
Rivadeneira, Marcela (2009): El voseo en medios de comunicación de Chile. Descripción y análisis de la
variación dialectal y funcional. PhD dissertation. Barcelona: Universitat Pompeu Fabra.
Stevenson, Jeffrey (2007): The sociolinguistic variables of Chilean voseo. PhD dissertation. Seattle:
University of Washington.
Torrejón, Alfredo (1986): “Acerca del Voseo Culto de Chile”, Hispania 69, 3, 677-83.
49
Use of Parallel Texts as a New Approach to Spatial Cognition:
The Case of English and Czech
Magdalena Honcova & Wei-lun Lu (Masaryk University Brno)
Prepositions have been a well-investigated area in cognitive linguistics, with already
quite a number of studies addressing the conceptual or constructional aspect of their
use (Lakoff 1987, Tyler and Evans 2003, among others). However, most of these studies
are based on either intuition or (a comparison of) mono-lingual corpora at the best. In
view of this gap, we propose an alternative methodology of investigating how speakers
of different languages verbalize their understanding of space in the same usage event.
We suggest to look at the case of English down and its Czech counterparts by using the
first chapter of Alice in Wonderland and its four Czech translations as our research
material.
One of the most striking finding that the parallel texts uncovered are discrepancies in
the distinction of static and dynamic terms in English and Czech (Dušková 2006), i.e.
whether the trajector in a conceptual scene is stationary or in motion. What is
perceived as dynamic in English is, in some cases, expressed by static preposition in
Czech. While English uses down to code both static and dynamic relations, in Czech we
observe a functional split between s (literally 'from'), po (literally 'onto') in dynamic
terms, and v (literally 'in') in static terms. Moreover, we observe that in Czech, a spatial
relationship may be realized by constructions at the morphological level (inflections in
particular), which serves the semantic function of a preposition. English, on the other
hand, does not take advantage of similar constructional means for verbalizing a spatial
relationship.
The study further revealed that discrepancies occur even in within-language
processing, as the four translations ranging from 1902 to 1961 expose a great variety in
coding the same concepts. This supports the notion that meaning is construed
dynamically, that is, the structures are "continually evolving" (Langacker 1987). This
study therefore also sheds light on how the degree of entrenchment rationally increases
prototypicality of symbolic units (Langacker 1987).
We conclude that our study reveals the radically conventional nature of grammar
(Croft 2001), especially how language utilizes available constructional means to code
spatial relationship. We in addition hope to show the usefulness of parallel texts as a
new methodology in the study of spatial cognition across languages.
50
References
Croft, William. (2001). Radical Construction Grammar. Oxford: OUP.
Dušková, Libuše. (2006). Mluvnice současné angličtiny na pozadí češtiny. Praha: Academia. Lakoff, G.,
& Johnson M. (1987/2003). Metaphors we live by. Chicago, US-IL: University of Chicago
press.
Tyler, Andrea, V. Evans. (2003).The semantics of English prepositions: Spatial scenes, embodied
meaning and cognition. Cambridge; New York: CUP.
Appendix
(1 ) Chlapec
The boy
slezl
climbed-3FIN
(2) Pluli jsme
We-NS sailed-1FIN
(3) Je nyní
Now-ADV she-NS is-3FIN
se
down-PREP
stromu. (Dynamic)
the tree.
po
down-PREP
proudu Temže. (Dynamic)
the Thames-OBL.
nízko
deep-ADV
na
společenském žebříčku. (Static)
down-PREP the social ladder-OBL.
51
The polysemy of the Spanish verb sentir: a Behavioral Profile
analysis
Marlies Jansegers, Clara Vanderschueren & Renata Enghels (Ghent University)
As for other linguistic fields, within the area of semantics several authors have been
arguing for the need to pursue corpus-linguistic methods in order to reduce the share of
subjective interpretation (Glynn 2010). In this talk, we propose to use the ‘behavioral
profile’ (BP) (Gries 2010) in order to disentangle the intricate polysemy of the Spanish
perception verb sentir (‘to feel’). Indeed, just like the more-studied ‘p rototypical’
perception verbs ver (‘to see’) and mirar (‘to look’), sentir shows an ample gamut of
semantic uses, in various syntactic environments, ranging from purely physical
perception (1), over cognitive perception (2) and emotive meanings (3), to a rather
discursive use (4):
1) Ella siente la mano de Tomasa.
‘She feels Tomasa’s hand.’
2) Para que sintieran que su poder es en sí mismo frágil.
‘So that they could sense that their power is in itself fragile.’
3) Indalecio sintió cierta ternura […] por aquella chica.
‘Indalecio felt a certain tenderness for that girl.’
4) Lo siento, señor, pero […] me encontraba un poco despistada.
‘ I am sorry, Sir, but I was a bit distracted’
The BP approach starts from a fine-grained manual annotation of a large number of
syntactic, morphological, semantic and pragmatic variables, called ID-tags.
Subsequently, the frequency data for each ID-tag per (predefined) semantic meaning are
processed through various kinds of statistical analyses (mainly hierarchical
agglomerative cluster analysis and correlational analysis). Thus, by gathering the
largest possible amount of distributional contextual information, a clear picture can be
made of the polysemy of a lexical item. In concrete terms, we annotated 1810 matches of
the lemma sentir, randomly selected from the Peninsular Spanish part of the CREAcorpus, for 138 ID tag levels. The BP analysis allows us to address the following
questions: (1) How are the different meanings structured: how many senses should be
distinguished – i.e. which meanings clust er together and which meanings should be
kept separately? (2) What is the prototype of each cluster of meanings? (3) Which
meanings are more related to each other and which are highly distinguishable? (4) What
variables make them more or less distinguishable?
52
The first results of a pilot cluster analysis show that two big significant meaning
clusters can be distinguished, which coincide with the division between the middle
voice uses (sentirse) and the other uses (sentir). Within these clusters, a number of
meaningful subclusters arise, which largely seem to coincide with the more general
semantic categories of physical, cognitive and emotional perception. Further analyses
will reveal how different these clusters are from each other, and which ID tags
prototypically characterize each of these clusters.
References
Glynn, Dylan. 2010a. Corpus-driven Cognitive Semantics. Introduction to the field. In Dylan
Glynn & Kerstin Fischer (eds.), Quantitative methods in cognitive semantics: corpus-driven
approaches, 1-41. Berlin & New York: Mouton de Gruyter.
Gries, Stefan Th. 2010. Behavioral Profiles: a fine-grained and quantitative approach in corpusbased lexical semantics. The Mental Lexicon 5(3). 323-346.
53
“WHAT? THAT!”Relative clauses as intersubjective
constructions in signed languages
Maria Josep Jarque (University of Barcelona) & Esther Pascual (University of
Groningen)
This paper stems from the assumption that language is modeled by the structure of
face-to-face interaction (Voloshinov 1929; Vygotsky 1962; DuBois 2001; Verhagen 2005;
Zlatev et al. 2008). The focus is on the interpersonal conversational pattern of question(answer), grammaticalized as a relative clause construction. Typological research has
shown that the wh-marker is one of the most common sources for relative markers
across the languages of the world (Sankoff and Brown 1976; Herring 1991; Heine and
Kuteva 2002; Lehmann 2008).
Since signed languages are discourse-oriented and used mostly in face-to-face
interaction, we suggest that this grammaticalization path is not only present in the
languages of the visual-gestural modality, but also constitutes the default strategy for
relativization. We show that in signed languages relative clause constructions display
similar features as information-seeking questions. Consider this example from Catalan
Sign Language, marked by two non-manual features: a browraise spread over the entire
nominal phrase, signaling topic, and, within its scope, a forwardhead tilt, specifically for
the relative. Both markings coincide with the non-manual markings for polar questions:
(1) [DEAF [OF LIVE IX-THERE)]headforward]browraise KNOW
Lit. ‘The deaf (man)? The one who was living there? He knew (about that possibility)’
‘The deaf man who was living there knew (about that possibility).’
A non-conversational scenario is presented through interactional means. The nongenuine question sets up a fictive interaction (Pascual 2002, in press) between a fictive
questioner and answerer, corresponding to the factual signer producing the utterance.
This investigation is based on a bibliographic study of 22 signed languages and a
qualitative analysis of a corpus of naturalistic Catalan Sign Language data. This dataset
includes monologues and conversations in various discourse modes (narrative,
descriptive, and informative) of different genres (from ordinary conversations to
television newscasts and videoblogs). The study contributes to the array of research
arguing for an interactional basis of linguistic structures, based on the primacy of
conversation (cf. Pascual 2002, in press).
54
References
Du Bois, J.W. 2001 [Forth.]. Towards a dialogic syntax. Special issue, ‘Dialogic
resonance’.Cognitive Linguistics.
Heine, B. & T. Kuteva. 2002. World Lexicon of Grammaticalization. Cambridge: CUP.
Herring, S.C. 1991.The grammaticalization of rhetorical questions in Tamil. In E.C. Traugott & B.
Heine (eds). Approaches to Grammaticalization. Amsterdam: Benjamins.
Lehmann, Ch. 2008. Information structure and grammaticalization. In M.J. López-Couso & E.
Seoane Posse (eds). Theoretical and Empirical Issues in Grammaticalization 3. Amsterdam:
Benjamins.
Pascual, E. 2002. Imaginary Trialogues: Conceptual Blending and Fictive Interaction in Criminal Courts.
Utrecht: LOT.
Pascual, E. In press. Fictive Interaction: The Conversation Frame in Thought, Language, and Discourse.
Amsterdam: Benjamins.
Sankoff, G. & P. Brown. 1976. The origins of syntax in discourse: A case study of Tok Pisin
relatives. Language 52: 631–666.
Verhagen, A. 2005.Constructions of Intersubjectivity: Discourse, Syntax, and Cognition. Oxford: OUP.
Vygotsky, L.S. 1962. Thought and Language.Cambridge: MITPress.
Voloshinov, V.N. [1929] 1986. Marxism and the Philosophy of Language. Cambridge:HUP.
Zlatev, J., et al. (eds.). 2008. The Shared Mind. Amsterdam: Benjamins.
55
Classroom Applications of Cognitive Linguistic Key Concepts
Dorothee Kohl-Dietrich, Constanze Juchem-Grundmann, Wolfgang Schnotz
(University of Koblenz-Landau)
Phrasal verbs (PVs) are considered to be a notoriously difficult field for learners of
English as a foreign language (EFL). Many lexemes that frequently occur in PVs (e.g. to
get, to come, to go as verbs and up, down, out as particles) are highly polysemous.
Furthermore, the idiomatic and opaque nature of many PVs can also be regarded as an
impediment for learners. Conventional teaching approaches to PVs have been rather
unsystematic, expecting students to simply learn PVs by rote.
Emphasizing that almost all conventional language use is conceptually motivated
(Langacker 2008), CL key concepts might serve pedagogical ends and help students
study PVs more systematically. It is assumed that the meaning of the particle
contributes significantly to the overall understanding of PVs and that even very opaque
PVs can be analysed on the basis of the meaning of the particle (Alejo 2010). According
to Tyler & Evans (2003) the different senses of a linguistic form build motivated
semantic networks which are distinct from but related to the so-called PROTOSCENE.
Thus, the more abstract senses of particles are related to their prototypical or basic
senses as prepositions (e.g., up = upwards movement => more, better, complete, to go up
= increase). Teaching students these semantic networks can endow them with strategies
to cope with the seeming arbitrariness and opacity of more figurative PVs.
Although a substantial body of CL-based research on PVs already exists (e.g., Condon
2008), empirical studies embedding CL-methods in a teaching and learning process in
order to prove the ecological validity of such an approach are still rather scarce.
This paper presents the results of a quasi-experimental field study with pre-/
posttest design (N=200-250, German EFL students, year 9, still ongoing), testing the
effect of CL-based instruction (experimental group) in comparison to conventional
teaching methods (control group). On the basis of the empirical data the following
research questions are discussed:
Q1:
Q2:
56
Is a CL-based teaching approach more efficient than conventional teaching
methods with regard to the retention of PVs under the condition that deep
cognitive processing is triggered in both groups?
Does a CL-based teaching approach facilitate transfer to novel PVs?
First pilot studies (N=81) were conducted with no significant effect with regard to Q1
(retention). Yet, concerning the understanding of new PVs (transfer), the CL-instructed
students outperformed the control group significantly (p < 0.05).
References
Alejo González, R. (2010). Making sense of phrasal verbs: A cognitive linguistic account of L2
learning. AILA Review, (23), 50–71.
Condon, N. (2008). How cognitive linguistic motivations influence the learning of phrasal verbs.
In F. Boers & S. Lindstromberg (Eds.), Cognitive Linguistic Approaches to Teaching
Vocabulary and Phraseology (pp. 133–158). Berlin, New York: Mouton de Gruyter.
Langacker, R. W. (2008). Cognitive grammar as a basis for language instruction. In P. Robinson &
N. C. Ellis (Eds.), Handbook of Cognitive Linguistics and Second Language Acquisition (pp. 66–
88). New York: Routledge.
Tyler, A. & Evans, V. (2003).The Semantics of English Prepositions: Spatial Scenes, Embodied Meaning,
and Cognition. Cambridge, New York: CUP.
57
A non-categorical approach to the grammatical functions of
gesture
Kasper Kok (VU University Amsterdam)
The functions of co-speech gestures in spoken discourse are not all that different
from those of speech. Like speech, gestures can present semantic content, mark
emphasis or personal commitment to the expressed content, structure the overall
organization of the discourse, and contribute to the perlocutionary force of an
utterance (Kendon, 2004; Müller et al., 2013a). These potentials have recently fueled an
upsurge of interest in the connections between gesture research and cognitivefunctional theories of grammar (Cienki, 2012; Muntigl, 2004; Turner and Steen 2012).
An important question that arises when considering how construction-based
grammar models can accommodate bodily movements, is whether their discourse
functions are linked with delineable form parameters. Recent research has addressed
this question in a form-first approach that aimed at identifying ‘recurrent’ gestural
patterns (Müller et al., 2013b). This approach revealed a number of clusters in gesture
form and context of occurrence. The observed variance within these patterns resonates
with a view of gestures as symbolic units of a schematic nature (Ladewig, 2011; Wilcox &
Xavier, 2013).
A possible drawback of this research is the assumption of a one-to-one mapping
between gesture categories and their functions. In actual spoken discourse, gestures
appear strongly multifunctional: in addition to representing semantic content, they
often provide emphasis to what is said and manage the interpersonal space between
interlocutors. In the current contribution, I pursue an approach to mapping out the
(grammatical) functions of gestures that does not assume one-to-one categorical
mappings.
This view is operationalized by means of a large-scale video survey. In an internetbased study, approximately 500 route-description gestures (derived from the Bielefeld
Speech and Gesture Alignment corpus; Lücking, Bergman et al. 2012), have been rated in
terms of a set of eleven possible functions (e.g., reference, ascription, emphasis,
hedging, meta-communicative) on a seven point likert scale, each by fifteen
independent raters. This results in a characterization of the different degrees to which
each of these gestures performs a range of different functions. Because the corpus has
also been annotated in terms of gesture form parameters (e.g. hand shape, orientation,
58
and movement), these data allow for a detailed scrutiny of the formal correlates of
gesture’s functions.
A second study addresses the question to what extent gestural meaning is
constrained by the verbal context. The exact same procedure was carried out, but the
videos were deprived of sound. This provides further insights into the contributions of
gestures to the discourse, above and beyond those of speech. I conclude by discussing
implications of this research for the notion of entrenchment and conventionalization
(Langacker, 1987). Accordingly, I consider possible extensions of grammar models
(Cognitive Grammar in particular) that are needed to account for the structure and
functions of co-verbal behaviors in their full import.
References
Cienki, A. (2012). Usage events of spoken language and the symbolic units we (may) abstract
from them. In J. Badio & K. Kosecki (eds.), Cognitive Processes in Language. Bern: Peter
Lang, 149-158.
Kendon, A. (2004). Gesture: Visible action as utterance. Cambridge University Press.
Langacker, R. W. (1987). Foundations of Cognitive Grammar: theoretical prerequisites. Volume 1 (Vol.
1). Stanford university press.
Ladewig, S. H. (2011). Putting the cyclic gesture on a cognitive basis.CogniTextes. Revue de
l’Association française de linguistique cognitive, (Volume 6).
Muntigl, P. (2004). Modelling multiple semiotic systems: The case of gesture and speech.
Perspectives on multimodality, 31-50.
Lücking, A., K. Bergman, et al. (2012). Data-based analysis of speech and gesture: The Bielefeld
Speech and Gesture Alignment Corpus (SaGA) and its applications. Journal on Multimodal
User Interfaces: 1-14.
Müller, C., S. H. Ladewig, et al. (2013). Gestures and speech from a linguistic perspective: A new
field and its history. In C. Müller, A. Cienki and E. Fricke et al. (eds) Body - Language Communication: An International Handbook on Multimodality in Human Interaction. Berlin,
Mouton de Gruyter.
Müller, C., J. Bressem, et al. (2013). Towards a grammar of gestures: A form-based view. In C.
Müller, A. Cienki and E. Fricke et al. (eds) Body - Language - Communication: An
International Handbook on Multimodality in Human Interaction. Berlin, Mouton de Gruyter.
Turner, M. and F. Steen (2012). Multimodal Construction Grammar. Available at SSRN 2168035.
Wilcox, S., & Xavier, A. N. (2013). A framework for unifying spoken language, signed language,
and gesture. Revista Todas as Letras, 15(1).
59
V1-Conditionals in German and English: a
Constructionalization Perspective
Torsten Leuschner (Ghent University)
In this contribution I propose a first constructional approach, from both a synchronic
and a diachronic point of view, to V1-conditionals like the following in German and
English:
1. Scheitert der Euro, dann scheitert Europa. (Angela Merkel)
2. Should the Euro fail, Europe will fail.
While such constructions are without doubt conditionals (Van den Nest 2010: ch. 2),
the initial position of the finite verb in the protasis is clearly reminiscent of polar
interrogatives. This suggests that V1-conditionals are a node in a constructional
network linking conditionals and interrogatives, or semantically speaking, that their
meaning can be analysed in terms of frame overlay (cf. Verhagen 2000 on concessives,
adopting ideas from Fauconnier/Turner 1998). Judging from the empirical data in
[anonymized], both is clearly the case in German but not in English, where V1conditionals are better described as a severely constrained conditional subschema
whose link with interrogatives is more apparent than real.
These observations in turn throw an interesting light on historical developments. In
this regard, the distinction between initial constructionalization on the one hand and
post-constructionalization changes on the other hand (Traugott/Trousdale 2013) proves
especially helpful. Philologists have often speculated on the rise of V1-protases from
discourse patterns involving polar interrogatives in early Germanic (e.g. Jespersen 1909:
376), i.e. on their constructionalization by recategorization from V1 main clauses just
prior to the onset of written documentation; by contrast, little attention has been paid
to subsequent constructional changes. Two points are worth making in this regard. On
the one hand, statistical data based on historical corpora ([anonymized]) suggest that
post-constructionalization changes have been a particularly protracted affair in
German, leading to different degrees of divergence from interrogatives and
specialization for conditionality at the levels of lexis, morphology and syntax in each
language. On the other hand, the data reinforce the need to view low-level changes as
embedded in changes higher up the constructional network, a point emphasized by
Traugott/Trousdale (2013) but often neglected by contrastive approaches (e.g. König
2012: 8-10 on V1-conditionals). Not only have English V1-conditionals lost any
60
productive link with interrogatives, their relative constructional isolation (as already
hinted by Hawkins 1986: 211) is in part a consequence of the fixation of SVO and the
attendant loss of paradigmatic oppositions between different word order options which
came to distinguish English from German early in its historical development.
References
Fauconnier, Gilles / Turner, Mark (1998): “Conceptual Integration Networks.” In: Cognitive
Science 22, 133-187.
Hawkins, John A. (1986): A Comparative Typology of English and Geman: Unifying the Con-trasts.
London: Routledge.
Jespersen, Otto (1909): A Modern English Grammar on Historical Principles. Part V: Syntax. Fourth
Volume. Heidelberg: Winter.
König, Ekkehard (2012): “Contrastive Linguistics and Language Comparison.” In: Languages in
Contrast 12, 3-26.
Traugott, Elizabeth Closs / Trousdale, Graeme (2013): Constructionalization and Construc-tional
Changes. Oxford: O.U.P.
Verhagen, Arie (2000): “Concession Implies Causality, Though in Some Other Space.” In: CouperKuhlen, Elizabeth / Kortmann, Bernd (eds.), Cause, Condition, Contrast: Cognitive and
Discourse Perspectives. Berlin/New York: Mouton de Gruyter, 361-380.
61
Iconicity and usage: a quantitative corpus-based study of
European causatives
Natalia Levshina (Université catholique de Louvain)
Introduction
The study focuses on variation in expressing causative events in fifteen IndoEuropean languages from the Germanic, Romance and Slavic groups. Based on a parallel
corpus of film subtitles, the study investigates the factors that determine the choice
between periphrastic and lexical causatives in a series of logistic regression analyses. It
thus sheds light on the relationships between frequency and iconicity as major factors
that shape up language structure.
Background
In functionally oriented studies, the ‘division of labour’ between periphrastic (1a) and
lexical causatives (1b) is traditionally explained by the iconicity principle: lexical
causatives convey more ‘direct’ causation with a stronger conceptual integration of the
cause and effect, whereas periphrastic causatives convey more indirect causation (e.g.
Haiman 1985).
(1) a. The sheriff caused Bob to die.
b. The sheriff killed Bob.
However, there exist other proposals, which focus on the role of usage. For example,
Haspelmath (2008) shows that frequently occurring events tend to have more basic,
unmarked forms, whereas rarer events are encoded with the help of more complex
forms. This study is an attempt to disentangle the iconicity and usage factors in
variation of causative constructions with the help of corpus-based multivariate
analyses.
Data and method
The data come from a self-compiled multilingual corpus of aligned film subtitles and
fiction. The data set contains approximately 700 multilingual exemplars that contain
periphrastic and lexical causatives (10,500 data points). The exemplars are coded for the
type of causative construction and a number of conceptual and usage variables, based
62
on previous theoretical accounts, e.g. agentivity of the Causee and volitionality of the
Causer. The frequencies are extracted from large comparable monolingual reference
corpora. A series of logistic regression models are fit with the above-mentioned
variables as predictors, and the type of the causative construction as the response, and
the 95% confidence intervals are compared (see an illustration in Figure 1).
Results
In accordance with the iconicity hypothesis, the variables that are associated with
the (in)directness of causation, behave in a similar way across the languages, as can be
seen from the 95% confidence intervals, which tend to overlap (see Figure 1). However,
adding the frequency of causation events as an independent variable significantly
improves the predictive power of the models, which suggests that both usage and
iconicity shape up language structure.
References
Haiman, J. (1985) Natural Syntax: Iconicity and Erosion. Cambridge University Press, Cambridge.
Haspelmath, M. (2008) Frequency vs. iconicity in explaining grammatical asymmetries.Cognitive
Linguistics, 19(1): 1–33.
63
Figure 1. Overlapping 95% confidence intervals of iconicity-related variables in eight
Germanic and Romance languages.
64
The “Belgian Tetris”: assessing the political impact of
metaphors on citizens’perception of and attitude towards
Belgian federalism.
Julien Perrez (Université de Liège) & Min Reuchamps (Université catholique de
Louvain)
In the literature, the political impact of metaphors has often been taken for granted
from metaphor analysis in political discourse, be it elite discourse or media discourse.
However, a more global understanding of what this political impact could consist of, is
still lacking from the current research agenda. As Koller (2009:121) puts it: “metaphor
helps construct particular aspects of reality and reproduce (or subvert) dominant
schemas.” To be able to account for how metaphors, through discourses, actively shape
the political reality, it is important to look at the relationships between metaphorical
discourses and their environment.
Based on the idea that metaphors do not only reflect the perceived reality, but also
function as cues through which citizens come to understand complex political processes
and through which they shape political behaviors, the aim of this study is precisely to
look at how specific metaphors might impact on the citizens’ framing of Belgian
federalism.
To measure the impact of metaphors on the citizens’ political representations and
attitudes, we developed an experimental set-up based on an article published in the
Belgian newspaper Le Soir (13-14 July 2013) in which Belgian federalism was deliberately
compared to a Tetris game. The original article included a picture and a text (208
words), which were used as authentic experimental material. For this experiment, we
distinguished three experimental conditions and one control condition. In the first
experimental condition (full condition), the participants were exposed to the original
article (including the text and the picture). In the second and third experimental
conditions, the participants were respectively exposed either to the text (text
condition) or the picture (picture condition). In the control condition, the participants
weren’t exposed to any metaphorical material at all. In the second stage of the
experiment, the participants were asked to achieve three interrelated tasks: (i) a free
description task, based on a free description of their own perception of Belgian
federalism, (ii) an association task, in which they had to select a picture which they
found the most appropriate to describe Belgian federalism, and finally (iii) a
questionnaire measuring the participants’ political knowledge of Belgian federalism and
attitudes towards its future development. In a post-test held four weeks after the first
65
experiment, the three tasks of the second stage have been replicated. This experiment
has been conducted in autumn 2013 among 400 students.
Comparing the various experimental conditions will make it possible (i) to measure
the impact of the Tetris metaphor on the citizens’ perceptions and representations of
Belgian federalism, (ii) to assess to what extent the different metaphorical media
differently contribute to this impact and (iii) to measure the long-term impact of this
metaphor on the citizens’ political representations and attitudes. In answering these
questions, this study will contribute to a better understanding of the role and functions
metaphors play in political discourse, and more globally in our everyday political
interactions.
66
Indefinite uno in Spanish: an exploratory study based on
two discursive genres
Marie Rasson (Université catholique de Louvain)
The aim of this paper is to analyze the discursive functioning of the Spanish pronoun
uno (‘one’/‘you’), in its non-numeral uses. Such uses are typically described in the
literature as "generic", "impersonal" or "undefined." In this study, I will show what
mechanisms and factors influence their interpretation, paying specific attention to
differences in discourse genre. Furthermore, I will propose an analysis of uno as strategy
of intersubjectivity.
I will establish a classification of undefined uno through a qualitative analysis of
occurrences obtained from two different corpora, namely an informal conversations
corpus (CORLEC) and a written academic discourse corpus (CLAE), which will allow me
to discover which factors influence the interpretative mechanisms. Using two corpora
of different genres/text types allows me to analyze the impact of the factor discourse
genre on the use of uno.
I propose a first typology of different uses in terms of reference of this pronoun. It
can vary from generic (1) to a particular individual, who can be the speaker (2), the
hearer (3) or another person (4). However, I also observe different degrees of genericity
within these categories, which justifies the qualitative analysis.
(1) Este ejemplo es el contrario de lo que sucede en poblaciones como Woodland California en
donde la ciudad es primordialmente mexicana y el español se puede escuchar en casi
cualquier lugar que uno visite. (CLAE)
‘This example is the opposite of what happens in places like Woodland California
where the city is primarily Mexican and Spanish can be heard in almost any place
you visit.’
(2) Si hoy se acabara el agua pues me agarraría de sorpresa ya que uno vive al día y para
planear lo que uno haría a futuro pues tendría uno que empezar a cambiar los hábitos.
(YCCQA)
‘If the water ran out today, it would take me by surprise since you live day by day
and to plan what you would do in the future, you would have to begin to change
habits.’
67
(3) Claro, cuando uno tiene ya muchas ganas de hacer algo... (CORLEC – CONV)
‘Of course, when you are already very eager to do something …’
(4) Tuvo muchos críticos Cantinflas, pero sin críticos, uno no puede tener carácter. (CLAE)
‘Cantinflas had many critics, but without critics, you cannot have character.’
I will show that some uses of this undefined uno are marks of intersubjectivity, as
defined by Benveniste (1966) and Traugott (2003), i.e. an expression of attention of the
speaker / writer to the hearer. Therefore I will propose a scale of intersubjectivity of the
uses of uno, in order to describe in more detail their functioning in both genres/text
types.
Corpora
CLAE: Corpus de lenguaje académico en español. 2009.
http://www.lenguajeacademico.info.
CORLEC: Corpus del Español centro-peninsular, 1990-1993.
References
Benveniste, Emile. 1966. « De la subjectivité dans le langage. » Problèmes de linguistique générale.
Paris: Editions Gallimard.
Traugott, Elizabeth Closs. 2003. “From subjectification to intersubjectification.” In Hickey,
Raymond (ed.), Motives for Language Change. Cambridge: Cambridge University Press,
124-139.
68
The epistemic large-quantity construction in Italian: [non so
quant_ + N/V] (I don’t know how many/much + N/V)
Valentina Russo (Humboldt University Berlin & Università degli studi di Napoli
"L'Orientale")
This paper deals with a poorly studied epistemic construction for expressing
indefinite quantity: Italian [non so quant+N(oun)/V(erb)] (I don’t know how many/much +
N/V). This construction is used to express large or exaggerated quantities in a
rhetorical way by conveying one’s own commitment toward it. It is argued that the
semi-idiomatic quantity construction (Antonopoulou/Nikiforidou, 2011) [neg-knowVerb
Q(quantifier)]Q [N/V] has distinct syntactic, semantic and pragmatic features of its own
that make it different from the homonym complementation construction [neg-know]Verb
[Q + N/V] Compl (s. Russo/Dittmar, 2014), as exemplified below:
[Non so]Verb [[quanti]Q [chili]NP [ha perso]VP]Compl
perso]VP
I do not know how many kg he has lost
>> [[Non so quanti]Q [chili]N]NP [ha
I know that he has lost many kilos,
probably more than I expected, but I
don’t know exactly how many
The paper raises the following questions: what makes the two constructions
different? Which morpho-syntactic, textual and prosodic constrains enable a
different segmentation of the “non so quant_” sequence, i.e. [neg-Know+Q+NP/VP]?
I will address these questions relying on the notion of constructions (Goldberg, 1995;
Croft, 2001), accounting for those aspects that characterize the epistemic semisubstantive quantity construction in oral and written usage. The analysis is based on
real corpus data, consisting of spontaneous oral interactions (LIP, Lablita), written
literature (DiaCoris) and Web occurrences. Results will bring evidence of a variety of
functional and formal features that apply to both oral and written uses, such as:
tense, mood and person of the verb know; mood of the following verb; syntactic
and semantic restrictions; informational status and other constrains. I will finally focus
on the prosodic profile of non so quant_ in the quantity construction, examining
pitch movements, integration with following or preceding elements and
phonological reduction (McGregor, 2013), which contribute to make it conventional
and psychologically entrenched in the speech community (Langacker, 2005: 140).
69
References
Antonopoulou, Eleni / Nikiforidou, Kiki (2011), Construction grammar and conventional
discourse: A constructionQbased approach to discoursal incongruity . In: Journal of
Pragmatics 43 (2011), 2594–2609.
Croft, William (2001), Radical Construction Grammar, Oxford: Oxford University Press.
DiaCORIS. http://dslo.unibo.it/coris_ita.html
Goldberg, Adele E. (1995), Constructions: A Construction Grammar approach to argument structure.
Chicago: Chicago University Press.
LABLITA: Cresti, Emanuela (2000), Lablita. Corpus di Italiano Parlato, voll. IQII, CDQROM.
Firenze: Accademia della Crusca.
Langacker, Ronald W. (2005). Construction grammars: Cognitive, radical, and less so . In:
Francisco J. Ruiz de Mendoza Ibanez & M. Sandra Pena Cervel (eds.). Cognitive
linguistics: Internal dynamics and interdisciplinary interaction. Berlin: de Gruyter, 101Q159.
LIP = DE MAURO, Tullio, MANCINI, Federico, VEDOVELLI, Massimo & VOGHERA, Miriam. 1993.
Lessico
di
frequenza
dell’italiano
parlato.
Milano:
Etaslibri
(http://www.parlaritaliano.it/index.php/visualizzaQcorpus?path=/Napoli/NA4&word
form=fine)
McGregor, William B. (2013), There are existential constructions and existential
constructions: PresunptionQinvoking existentials in English , in Folia Linguistica
(47/1), 139Q181.
Russo, Valentina / Dittmar, Norbert (in press), Konstruktionen konversationeller Vagheit
in deutschen und italienischen Gesprächen . In Selig / Dittmar / Morlicchio (eds.),
Gesprächsanalyse zwischen Syntax und Pragmatik. Deutsche und italienische Konstruktionen,
Tübingen: Stauffenburg Verlag.
70
Fictive interaction in the Hebrew Bible
Sergeiy Sandler (Ben-Gurion University)
Modern public discourse and the speech of contemporary youth appear to be
particularly ‘conversational’, regularly using interactional structures such as the like
construction (“It was like Why not?”) (Fairclough 1994; Streeck 2002). While
acknowledging that some specific conversational constructions may be novel or occur
more frequently in certain contexts, I suggest that fictive interaction—the use of the
basic frame of the conversation in order to structure language and discourse—is
overarching and widespread across genres and sociolinguistic groups (Pascual 2006,
forth.).
To explore this, I study an ancient and extremely influential religious text, the
Hebrew bible. The Hebrew bible shows a highly conversational structure, in the vast
occurrence of: (i) non-information seeking questions; (ii) the verbal root ‘‫( ’רמא‬amar), ‘to
say’; and (iii) direct rather than indirect speech (Rendsburg 1990; Miller 2003). I will
discuss examples of such frequent structures as: (i) ‘conversations’ with one’s mind or
parts of one’s body (e.g. “God said to his heart,…”); and (ii) the presentation of nonreported speech—ascribed to God, a person or group, and even an action—in order to
introduce actual or putative intentions, hopes, motives, or states of affairs (e.g. “Saul
blew the trumpet throughout all the land, to say let the Hebrews hear”, 1Sam 13:3). Special
emphasis will be laid on the grammaticalized form of the complementizer ‘‫( ’רמאל‬lemor),
from the infinitive of the speaking verb ‘‫ ( ’רמא‬amar) (Miller 2003). This complementizer
introduces the reason or significance of a particular action, presented through direct
speech (e.g. “And the officers […] were beaten saying [lemor] Wherefore have ye not…?”,
Exod 5:14).
By looking at Biblical Hebrew, we hope to shed some light on the hypothesis that
structuring language and discourse through the conversation frame is a non-trivial
overarching phenomenon, which I suggest reflects the dialogic nature of language.
References
Fairclough, N. 1994. Conversationalisation of public discourse and the authority of the
consumer. In R. Keat, N. Whitely & N. Abercrombie (eds.). The Authority of the Consumer.
London: Routledge.
Miller, C.L. 2003. The Representation of Speech in Biblical Hebrew Narrative: A Linguistic Analysis.
Winona Lake, IN: Eisenbrauns.
Pascual, E. 2006. Fictive interaction within the sentence. Cognitive Linguistics 17(2): 245–267.
71
Pascual, E. Forth. Fictive Interaction: The Conversation Frame in Thought, Language, and Discourse.
Amsterdam: John Benjamins.
Pascual, E. & S. Sandler (eds.). In prep. The Conversation Frame: Forms and Functions of Fictive
Interaction.
Rendsburg, G.A. 1990. Diglossia in Ancient Hebrew. New Haven, CT: American Oriental Society.
Streeck, J. 2002. Grammars, words, and embodied meanings. Journal of Communication 52(3): 581–
596.
72
What can gestures tell us about particle meanings?
Steven Schoonjans (University of Leuven)
A recurrent issue in the analysis of modal particles is the description of their
meaning. Indeed, the meaning of modal particles is highly context-dependent, which
makes it hard to offer a general meaning description. Closely related to this issue is the
fact that there is discussion about the exact relation between particles with resembling
meanings. Replacement tests and translation analyses have been brought in in attempts
to come to better grips with the particle meanings, but so far no unanimity has been
reached in these discussions.
In this talk, it will be shown that gesture research can shed a new light on the
meaning-related issues. This will be illustrated by means of a comparison of the gesture
patterns co-occurring with seven German modal particles, distributed over three
onomasiologically defined clusters: particles linking questions to the context (denn,
eigentlich), particles underlining the truth of the message in assertions (ja, doch), and
particles marking obviousness or self-evidence (eben, einfach, halt), as illustrated in the
examples below (taken from Thurmair 1989). The analysis is based on ca 21 hours of
video materials stemming from three different settings (talk shows, televised sports
reports, and parliamentary speeches).
1) Willst du morgen mitkommen zum Baden? – Hast du denn zur Zeit Urlaub?
'Would you like to join in for a swim tomorrow? – Do you denn have a holiday at the
moment?'
2) Im Auto hat man wenigstens freie Sicht, da hat man ja einen Scheibenwischer.
'At least in the car you have an unhampered view; there you have ja a windscreen
wiper.'
3) Warum sind die Frauen so hinter dir her? – Ich bin eben unwiderstehlich.
'Why are women so after you? – I am eben irresistible.'
Issues to be discussed in particular include the polysemy of the particle ja and the
relation of the particles eben, einfach, and halt. Being the most frequent modal particle of
German, ja is notoriously polysemous, which has lead to considerable disagreement
about how its meaning is to be described (see Rinas 2007). It will be shown how the
gesture distribution can contribute to mapping out this polysemy. As for eben, einfach
and halt, scholars disagree about the precise relation between these particles. Especially
73
the relation between eben and halt is controversial: Scholars such as Autenrieth (2002)
think they are synonymous, while others follow Thurmair (1989) in claiming that they
are not. Similarities and differences in the gestures patterning with these particles will
be referred to in order to arrive at a more nuanced view on the relation between the
particles.
References
Autenrieth, Tanja. 2002. Heterosemie und Grammatikalisierung bei Modalpartikeln. Tübingen:
Max Niemeyer.
Rinas, Karsten. 2007. "Bekanntheit? Begründung? Einigkeit? Zur semantischen Analyse der
Abtönungspartikel ja." Deutsch als Fremdsprache 44,4. 205-211.
Thurmair, Maria. 1989. Modalpartikeln und ihre Kombinationen. Tübingen: Max Niemeyer.
74
From reportative evidentiality to cohesion: a corpus study
between stance and discourse reference
Stef Spronck (University of Leuven/The Australian National University)
Reported speech is an evidential strategy. In constructing an utterance such as (1)
(1) ‘John said: “There is still hope"’
the current speaker (typically) invokes a discourse situation previous to the current
discourse situation involving the reported speaker John and the reported message.
Whether the reported message is relevant at the speech moment depends on the way in
which the reported speech situation and its participants relate to the current speech
situation. This opposition of two discourse events is the de ning feature the grammatical category of evidentiality (Jakobson, 1957). In deciding whether John's hopeful
assessment in (1) is still relevant at the speech moment and for the discourse participants involved, interpreting the evidential meaning is key.
Typologically, languages have been found to use direct speech constructions mostly
for functions that include an evidential meaning but in exceptional {but not
uncommon{ cases, examples have been attested of direct speech constructions being
used for func-tions beyond the evidential domain (see Pascual, fc: ch. 4). Functions of
direct speech constructions that include an evidential meaning are those that involve
some attri-bution of speech, thought or intentions to some discourse participant other
than the current speaker at the present time and place. A language that uses one single
con-struction to express all three of these functions is the Australian Aboriginal
language Ungarinyin. Attested functions of direct speech constructions that do not
include an evidential meaning have been as diverse as `causality' (voice), `lest'
(modality), `be-ginning of the action' (aspect), `future' (tense) and others. The Usan
example in (2) illustrates the second of these functions:
(2)
mi
qei-qei
mani umer-iner qamb gitab
ig-oun
thing some-RED yam wilt-3s:UF say:SS abstain:SS be-1p:PR
`We abstain from various things lest the yams wilt' (Reesink, 1993: 222)
In this paper I begin by showing that the discourse referential properties of direct
speech constructions with and those without or with a more non-prototypical evidential meaning, di er considerably. Since `evidential' reported speech constructions
75
(such as in Ungarinyin) involve two opposed discourse situations with their own inherent time, place and participants, reference to these discourse participants, places
and times displays two separate patterns: that of the current speech situation and that
of the reported speech situation. In non-evidential reported speech constructions this
distinction is absent, which is re ected in the discourse status of the discourse
participants referred to in a construction as in (2).
I then present a corpus analysis of Russian texts demonstrating that several types of
reported speech constructions in the language can be identi ed as contrasting evidential strategies and that these also display speci c patterns of discourse reference.
These patterns reveal important clues as to how the study of stance (Du Bois, 2007) and
discourse reference within a cognitive approach (Kibrik, 2011) may be combined. I
conclude by arguing that the study of reported speech constructions in connection to
discourse organisation o ers a perspective onto two of the most central cognitive
processes in constructing language: social cognition and memory.
References
Du Bois, J. W. (2007). The stance triangle. In R. Englebretson (Ed.), Stancetaking in Discourse, pp.
139-182. Amsterdam/Philadelphia: John Benjamins.
Jakobson, R. (1957). Shifters, verbal categories and the Russian verb. Department of Slavic Languages
and Literatures, Cambridge, MA: Harvard University.
Kibrik, A. A. (2011). Reference in Discourse. Oxford etc.: Oxford University Press.
Pascual, E. (fc.). Fictive Interaction: The Conversation Frame in Thought, Language, and Discourse.
Amsterdam/Philadelphia: John Benjamins.
Reesink, G. P. (1993). `Inner speech' in Papuan languages. Language and Linguistics in Melanesia 24,
217-225.
76
Humorous implications and meanings: a study of sarcasm in
interactional humor
Sabina Tabacaru (Université Lille 3 & University of Leuven)
According to Bergen and Binstead (2001), humor is “the least understood of our
cognitive capacities”, and since humor abuses inferences through linguistic imagery, it
should be cognitive oriented. My talk aims at defining the relationship between humor
and the human mind, as already pointed out in Cognitive Linguistics (Vandaele 2002;
Feyaerts 2004; Veale et al. 2006; Ritchie 2006; Brône 2008; Brône and Feyaerts 2003). More
specifically, I focus on underlining the importance of addressing sarcasm from the
perspective of the linguistic mechanisms creating it. I base my results on a large corpus
of examples drawn from two American TV-series: House M.D. and The Big Bang Theory.
The examples are annotated using ELAN which allows a more fine-grained analysis of
the data. Based on these results, I present a typology of sarcasm in the form of complex
cognitive phenomena, ranging from metonymy and metaphor to figure-ground
reversal. These sarcastic utterances are analyzed against the background of Clark’s
(1996) layering model and Fauconnier’s (1984, 1994) mental spaces theory, as suggested
by Brône’s (2008) unified account of humor, to show how the human mind creates and
understands sarcasm. I show that this unified account can also be used to explain
sarcastic utterances because it points out the importance of separating the discourse
base space from the pretense space in which sarcasm takes place. The discourse base
space represents the common ground between interlocutors, because the speakers must
be confident enough of all the assumptions the hearers are most likely to draw in
interaction. Given all the rich implications and meanings that it generates, I argue that
sarcasm is a fundamental means of all types of meaning construction (Radden et al.
2007).
References
Bergen, Benjamin and Kim Binsted. 2003. The Cognitive Linguistics of Scalar Humor. In Michel
Achard and Suzanne Kemmer (Eds.) Language, Culture, and Mind. Stanford: CSLI.
Brône, Geert. 2008. Hyper and misunderstanding in interactional humor. Journal of Pragmatics,
40: 2027-2061.
Brône, Geert and Kurt Feyaerts. 2003. The cognitive linguistics of incongruity resolution:
Marked reference-point structures in humor. University of Leuven, Department of
Linguistics preprint no. 205.
Clark, Herbert H. 1996. Using language. Cambridge: Cambridge University Press.
Fauconnier, Gilles. 1984. Espaces mentaux. Aspects de la construction du sens dans les langues
naturelles. Paris: Les Editions de minuit.
77
Fauconnier, Gilles. 1994. Mental spaces. Aspects of meaning construction in natural language.
Cambridge: Cambridge University Press.
Radden, Günter, Klaus-Michael Köpcke, Thomas Berg, and Peter Siemund. 2007. Introduction.
The construction of meaning in language. In Aspects of meaning construction, Radden
Günter, Köpcke Klaus-Michael, Berg Thomas, and Siemund Peter (Eds.): 1-15.
Amsterdam & Philadelphia: John Benjamins Publishing Company.
Ritchie, Graeme. 2006. Reinterpretation and viewpoints. Humor: The International Journal of Humor
Research 19: 251-270.
Vandaele, Jeroen. 2002. Humor mechanisms in film comedy: Incongruity and Superiority. Poetics
Today, 23(2): 221-249.
Veale, Tony, Kurt Feyaerts and Geert Brône. 2006. The cognitive mechanisms of adversarial
humor. Humor: International Journal of Humor Research, 19(3): 305-338.
78
Raised eyebrows as gestural triggers in interactional humor
Sabina Tabacaru & Maarten Lemmens (Université de Lille 3)
The growing interest in humor within the field of Cognitive Linguistics these past few
years (Brône and Feyaerts 2003; Veale et al. 2006; Brône 2008) shows that humor exploits
inferences through linguistic imagery and that it is highly creative. Following Yus (2003:
1299), we assume that humor uses discourse markers that allow the audience to see that
what is being said should not be taken seriously. In this study, based on a large corpus of
examples extracted from two American television series (House M.D. and The Big Bang
Theory), we add a yet unexplored multimodal perspective, that of facial expressions
accompanying humorous utterances. More specifically, we focus on raised eyebrows
used with sarcasm and hyper-understanding. We present a qualitative and quantitative
analysis of raised eyebrows used in interactional humor, arguing they play a role in the
switch of the context to a humorous interpretation, contributing in this way to meaning
construction. Our study analyzes these humorous utterances against the background of
Clark’s (1996) layering model and Fauconnier’s (1984, 1994) mental spaces theory (cf.
Brône 2008). We show how raised eyebrows function as “gestural triggers” allowing the
hearer to make the connection between explicature (i.e., what is explicitly
communicated by an utterance; cf. Carston 2002, 2004) and implicature (assumptions
that are not explicit and that the hearer has to infer from the contextual environment,
cf. Grice 1989). As such, we show that raised eyebrows play an important role in the
understanding of the humorous message, because they guide the hearer to interpret
utterances in a humorous way and they contribute to meaning construction.
References
Brône, Geert. 2008. Hyper and misunderstanding in interactional humor. Journal of Pragmatics,
40: 2027-2061.
Brône, Geert and Kurt Feyaerts. 2003. The cognitive linguistics of incongruity resolution:
Marked reference-point structures in humor. University of Leuven, Department of
Linguistics preprint no. 205.
Carston, Robyn. 2002. Thoughts and utterances: the pragmatics of explicit communication. Oxford:
Blackwell.
Carston, Robyn. 2004. Relevance Theory and the Saying/Implicating Distinction. In: Horn
Laurence R. and Gregory L. Ward (Eds.): 633-656, The Handbook of Pragmatics. Oxford:
Blackwell.
Clark, Herbert H. 1996. Using language. Cambridge: Cambridge University Press.
Fauconnier, Gilles. 1984. Espaces mentaux. Aspects de la construction du sens dans les langues
naturelles. Paris: Les Editions de minuit.
Fauconnier, Gilles. 1994. Mental spaces. Aspects of meaning construction in natural language.
Cambridge: Cambridge University Press.
79
Grice, Paul. 1989. Studies in the way of words. Cambridge: Harvard University Press. Veale, Tony,
Kurt Feyaerts and Geert Brône. 2006. The cognitive mechanisms of adversarial humor.
Humor: International Journal of Humor Research, 19(3): 305-338.
Yus, Francisco. 2003. Humor and the search for relevance. Journal of Pragmatics, 35:1295-1331.
80
Discourse position and polyfunctionality of the Spanish
markers anda, vamos, vaya and venga
Sanne Tanghe (Ghent University)
Major works on (discourse) markers derived from motion verbs have concentrated
mainly on their pragmatic functions (Company Company 2004, Romero Aguilera 2006,
Zorraquino y Portolés 1999). Nevertheless, recent studies have emphasized the
importance of taking into account also formal parameters for a more complete
description (Pons Bordería 2006, Melón Eva 2013, Pons Bordería and Estellés 2013).
Therefore this study focusses on the discourse position of the Spanish markers anda,
vamos, vaya and venga. The goal is to study whether there is a correlation between the
discourse position and the function of a marker and to what extent the position restricts
or defines the activation of a function. The level of mobility of each marker will be
measured in order to verify if greater mobility involves greater functional variability
(Pons Bordería 2010, 336).
The dataset consists out of transcriptions of oral corpora (CREA, MC-NLCH, COLam,
Val.es.co, Corlec) with a total of about 3100 occurrences. The framework for the analysis
of the discourse positions is the Val.es.co. model of discourse units (cf. selected
bibliography). This model takes into account the discourse unit formed by the marker
(intervention, act or subact) and the distribution of the unit within the discourse (initial,
medial, final or independent position of intervention, act or subact). Next, the occurrences of
the markers are classified according to their pragmatic values (appellative, expressive,
metadiscursive) and subvalues (surprise, disagreement, reformulation, etc.). The two variables
(position and function) will then be statistically tested for correlations (by applying a
Fisher-Freeman-Halton-test) and analyzed qualitatively.
A first approximation to the data shows us that we can expect there to be at least
some relation between both variables in that, for example, appellative or expressive
uses can be used independently as an intervention (1), while markers with a
metadiscursive function will be imbedded into a greater unit (2):
(1) <H6>Quería hacer una pregunta.
<H1> Venga.
<H6> Es que mi marido no quiere hacer testamento. (CORLEC)
81
(2) NM: #y / { SA bueno SA } / crees que ha cambiado la vida en el centro / desde que
han llegado todos estos muchachos / { SA vamos SA } de [/] de origen extranjero ?#
(C-ORAL-ROM)
The results of the quantitative study will provide a more detailed overview of the
mapping of meaning (pragmatic values) and form (discourse position) of the four
markers.
References
Briz and Pons Bordería (2010): Unidades, marcadores discursivos y posición. Los estudios de
marcadores del discurso en español, hoy. Loureda Lamas and Acín Villa. Madrid
Arco/Libros.
Briz and Val.Es.Co. research group (2003): Un sistema de unidades para el estudio del lenguaje
coloquial. Oralia 6: 7-61.
Cabedo Nebot (forthcoming): On the delimitation of discursive units in colloquial Spanish: Val.Es.Co
application model, John Benjamins.
Hidalgo (2006): Reconocimiento de unidades en el discurso oral: El acto y el subacto como segmentos
menores del análisis. Filología y lingüística: Estudios ofrecidos a Antonio Quilis vol.2,
César Hernández (ed.), 1541-1560. Valladolid: CSIC/UNED/Universidad de Valladolid.
82
Motor concepts, metaphor and gesture: The comparison
between referential gestures referring to concrete and
metaphorical motor concepts
Yao Tong & Alan Cienki (VU University Amsterdam/Moscow State Linguistic
University)
Some consensus has been reached in Cognitive Linguistics that much of our thought
is grounded in physical actions and meaning is in some ways embodied (Lakoff &
Johnson, 1999; Barsalou, 2008). Some motor concepts have been argued as image
schematic and as fundamental embodied experiences which give rise to more abstract
concepts (Johnson, 1987). Since speech and gesture can both serve as “windows into
thought,” the starting point of this research is how a motor concept as a minimal
analytic unit of thought (McNeill’s [1992] notion of the “growth point” of an idea) is
realized in linguistic and gestural modalities in natural discourse.
This research intends to investigate the conceptualization of motion in its
metaphorical nature, represented in visible/gestural forms of expression, from a
quantitative perspective. The metaphorical nature of a motor concept, i.e., (conceptual)
motor metaphor in this research, is regarded as a degree of metaphoricity ranging from
concrete to metaphorical conceptualization. This research intends to answer the
following research question: Where do differences lie in the referential gestures
referring to concrete and metaphorical motor concepts in terms of their amount and
function? The answer to this question could tell us more about the degree to which we
conceptualize physical and abstract motion with the same kinds of motor imagery.
As a quantitative study, this research aims to analyze about 300 referential gestures
in 10 TED talks (about 150 minutes) with the help of an automatic semantic annotation
web interface-- Wmatrix (Rayson, 2009). This program automatically assigns 21
semantic fields (e.g., education, emotions) to words of a text and has been found useful
in metaphor identification and analysis (Koller et al., 2008). The operational definition
of a motor concept is regarded as semantic field M “MOVEMENT, LOCATION, TRAVEL &
TRANSPORT.” The motor concepts related to cases of concordances detected by
Wmatrix are firstly examined and coded as concrete versus metaphorical and then
correlated referential gestures are picked out. The gesture coding involves the coding of
its form and function using the software ELAN. Forms will be coded in terms of gesture
unit, gesture phase (preparation, stroke, retraction) (Kendon 2004). Functions will be
coded based on Müller (1998) and Streeck (2008; 2009), including the following methods
83
of referring -- pointing, modeling, drawing, bounding or measuring, shaping, handling,
and acting. The expected result is that there are about same amount of referential
gestures referring to concrete and metaphorical motor concepts.
References
Barsalou, Lawrence W. 2008. Grounded cognition.
Johnson, Mark. 1987. The body in the mind.
Kendon, Adam. 2004. Gesture: Visible action as utterance.
Koller, Veronika et al. 2008. Using a semantic annotation tool for the analysis of metaphor in
discourse.
Lakoff, George, & Johnson, Mark. 1999. Philosophy in the flesh.
McNeill, David. 1992. Hand and mind. Müller, Cornelia. 1998. Iconicity and gesture.
Rayson, Paul. 2009. Wmatrix: a web-based corpus processing environment. Streeck, Jürgen.
2008. Depicting by gesture.
Streeck, Jürgen. 2009. Gesturecraft.
84
Conceptualisation-driven Machine Translation: a Case Study
for the Dutch and English Verb Phrase
Paul Van Eecke (Sony Computer Science Laboratory Paris)
Despite major improvements in the last several years, Machine Translation (MT) is
still far from perfect. State of the art statistical MT systems fail to correctly translate
even simple sentences, such as the one presented in (1).
(1) a. He has been playing . (Original English sentence)
b. Hij speelt . (Dutch translation by Google Translate1)
He plays .
While the translation correctly captures the lexical content, it fails to grasp the
temporal and aspectual information. This failure cannot be attributed to a lack of data
in the training corpus, because this kind of verb phrase is not infrequent at all. The
problem is much more fundamental, namely that the language models in current MT
systems are not sophisticated enough to handle more complex linguistic phenomena.
In this presentation, I will argue that insights from cognitive linguistics should be
integrated into MT systems in order to develop superior language technologies. More
specifically, language models need to be expanded with the rich conceptualisations that
are expressed by natural language grammars. I will illustrate this approach through a
case study on the English and Dutch verb phrase (VP). Concretely, I have implemented
computational processing models of the Dutch and English VP in Fluid Construction
Grammar (FCG) (Steels, 2011, 2012) that work for both parsing and production. The
models include a reconstruction of the conceptualisations that English and Dutch
speakers employ for situating events in time. The grammars consist of an automatically
generated lexicon of the 80 most frequent verbs in English and Dutch, and seven
grammatical tense-aspect-modality constructions.
Instead of using a form-to-form translation based on sheer statistics, the model
performs a deep semantic analysis of input sentences and extracts their conceptual
structure. This conceptualisation is then used for producing the translation in the target
language. I will show that this conceptualisation-driven approach not only outperforms
the form-based approach for correctly translating closely related tense-aspect-modality
1
Google Translate (http://translate.google.com). Consulted on 13/05/2014.
85
distinctions, but that it also successfully deals with cases in which Dutch and English use
completely different constructions (e.g. the progressive tense).
The contribution of my work is twofold. On the one hand, it provides an argument for
the incor-poration of more linguistic knowledge into machine translation. On the other
hand, it contributes to the understanding of the Dutch and English tense-aspectmodality systems, by providing an operational formalisation in Fluid Construction
Grammar.
References
Steels, L., editor (2011). Design Patterns in Fluid Construction Grammar. John Benjamins,
Amsterdam.
Steels, L., editor (2012). Computational Issues in Fluid Construction Grammar. Springer Verlag,
Berlin/Heidelberg.
86
Viewpoint representation and legitimization in Dutch
journalistic crime narratives
Kobie van Krieken & José Sanders (Radboud University Nijmegen)
Introduction and research questions
Journalists are storytellers. In the case of criminal acts, they write narratives about
the experiences of people involved in these acts; most often eyewitnesses, since they are
the most relevant journalistic sources.
The objectification and legitimization of source material has not always been a
professional demand, but has grown essential in the last decades. In our paper, we argue
that a cognitive linguistic approach should be adopted to adequately describe how
journalists have developed linguistic strategies to represent criminal acts from the
viewpoints of eyewitnesses without crossing the boundaries of nonfiction.
RQ1
What linguistic strategies are used in journalistic narratives to describe
criminal acts from the viewpoints of eyewitnesses?
RQ2
What linguistic strategies are used to legitimize these viewpoint
representations?
RQ3
How did viewpoint representation strategies and legitimization strategies
develop in the last 1.5 century?
Methodology
Building on a framework of mental spaces and blended spaces (Fauconnier, 1985;
Fauconnier & Turner, 2002), we developed a model for the cognitive linguistic analysis
of journalistic narratives, elaborating on Dancygier’s (2012) model of narrative fiction
and Author’s (2010) model of news narrative. This model enables the analysis of
linguistic strategies that embed and blend viewpoints in news narratives and accounts
for the genre-specific conventions of factuality and legitimization. We applied our
model to Dutch journalistic crime narratives that have been published between 1860
and 2010. For each decade within this period, ten narratives were selected and analyzed.
87
Results
In news narratives from all periods, verbs of perception and cognition are journalists’
main instruments to embed the viewpoints of eyewitnesses and relate crimes through
the experiences of these persons. Only in the most recent periods, present tense
narration of cognition and perception is employed as a strategy to blend the viewpoints
of eyewitnesses with the journalist’s viewpoint. This strategy provides an account of
criminal acts through the “on-line” thoughts and perceptions of eyewitnesses.
The representation of eyewitnesses’ viewpoints through space embedding and
blending dramatizes and “fictionalizes” the narrati ves, which calls for legitimization.
We argue that only in recent decades, a Narrative-External Discourse Space is
represented in journalistic narratives to warrant the truthfulness of the narrative by
grounding the exchange of information between journalist and source.
Conclusion
This study shows how journalists strategically use language to describe criminal acts
from the viewpoints of eyewitnesses, without disregarding the genre’s demands of
factuality and grounding. While strategies to embed and blend viewpoints have been
employed since the nineteenth century, more recently new blending techniques as well
as external discourse spaces have been developed that both dramatize and legitimize
narrative reconstructions of reality in an ever more sophisticated way.
References
Author (2010).
Dancygier, B. (2012). The Language of Stories: A Cognitive Approach. Cambridge: Cambridge
University Press.
Fauconnier, G. (1985). Mental Spaces: Aspects of Meaning Construction in NaturalLanguage.
Cambridge: Cambridge University Press.
Fauconnier, G., & Turner, M. (2002). The Way we Think: Conceptual Blending and the Mind's Hidden
Complexities. New York: Basic Books.
88
Chopping down the Syntax Tree
Remi van Trijp (Sony Computer Science Laboratory Paris)
Many linguists either explicitly or implicitly assume a tree structure (or phrase
structure) when analyzing linguistic utterances. As a data structure, trees consist of
nodes that have at most one parent node (e.g. in the phrase structure rule NP → DET N),
which means that information in a tree can only trickle down from a parent to its
immediate children, or percolate upwards in the other direction.
A tree structure is thus hopelessly inadequate for representing dependencies
between nodes that are in the top of the hierarchy and nodes that are situated
somewhere below. This problem becomes particularly apparent in long-distance
dependencies, as found in topicalization patterns (e.g. Ice cream I like!) or WH-questions
(e.g. What did you see?), where tree-based analyses require a lot of complex formal
machinery in order to respect the locality constraint of a tree structure. The most
common solution is the filler- gap approach, pioneered by Gazdar (1981), which requires
tree steps: (1) identifying a “gap” at an “extraction site”, (2) communicating information
about the gap node-by-node upwards in the tree, and (3) stopping the gap when a
“filler” has been found; as in WhatFILLER did you see___GAP?
In recent years, however, a cognitive-functional (or usage-based) alternative has
started to crystallize in which long- distance dependencies emerge spontaneously as a
side-effect of how constructions may freely combine with each other in order to cater
for the communicative needs of language users (Goldberg, 2002). Unfortunately, this
view has been criticized for being ad-hoc and not lending itself to scientific
formalization. The goal of this presentation is therefore to demonstrate that the
cognitive-functional alternative can be made formally explicit through a computational
model of English long-distance dependencies that works for both parsing and
production, implemented in Fluid Construction Grammar (Steels 2011). Moreover, the
implementation shows that a cognitive-functional approach outperforms the filler-gap
analysis in terms of theoretical parsimony (no special rules), completeness (competence
AND processing), and empirical adequacy (particularly for explaining language change).
More specifically, I will show that rather than directly “reading off” all the
information about an utterance from its tree structure, it is necessary to chop down the
syntax tree and treat different “linguistic perspectives” on equal footing as is common
in functional linguistics (Nuyts 2011). These perspectives include an utterance’s
functional structure, information structure and illocutionary force. I will show that
89
construction grammar is especially adequate for representing the complex interaction
between these perspectives, as constructions are capable of cutting across different
levels of linguistic organization.
References
Gazdar, Gerald (1981). Unbounded dependencies and coordinate structure. Linguistic Inquiry
(12):155—184.
Goldberg, Adele E. (2002). Surface generalizations: an alternative to alternations. Cognitive
Linguistics (13)4:327—356.
Nuyts , Jan ( 2011 ). Pattern versus process concepts of grammar and mind: a cognitivefunctional perspective. In: Mario Brdar , Stefan Th. Gries & Milena Žic Fuchs (Eds.),
Cognitive linguistics: convergence and expansion (pp−66.47). Amsterdam : John Benjamins .
Steels, Luc (ed., 2011). Design Patterns in Fluid Construction Grammar. Amsterdam: John Benjamins.
90
The (in)stability of familiarity judgments: individual variation
and the invariant bigger pictures
Véronique Verhagen & Maria Mos (Tilburg University)
Metalinguistic judgments are often used in linguistic research. In order to provide
judgments, people must draw on their (implicit) linguistic knowledge. Therefore,
judgments can be used as evidence for making inferences about people’s linguistic
representations. It is even argued that judgment data play a crucial role in linguistic
investigation because they provide quantitative information not readily available from
other kinds of data (Schütze & Sprouse, 2013). The reliability of such judgments,
however, is regularly called into question. We have conducted two studies investigating
the test-retest reliability of metalinguistic judgments, more specifically familiarity
judgments, for Dutch prepositional phrases (e.g. op de bank, ‘on the couch’). Importantly,
from a usage-based perspective, variation in judgments between and within speakers
might be expected: with judgments based in representations, and representations
resulting from input and use, both inter- and intra-individual variation are likely.
In the first study, familiarity judgments on 44 PPs offered in isolation and in sentence
context were given by 86 participants —all native speakers of Dutch— in two identical
test sessions using Magnitude Estimation. Our study looks at the stability and reliability
of these judgments in three ways: by investigating the test-retest reliabilities, by looking
at the effects of the presence and absence of context, and by examining to what extent
judgments can be predicted from stimuli’s frequency.
Aggregated scores (averaged over participants) are remarkably consistent (Pearson’s
r = .97), and strongly predicted by corpus frequencies (R 2 = .55). At the same time, there
is considerable variation between and within participants. Judgments on items vary
considerably from participant to participant. Furthermore, none of the individual
participants gave judgments during the two test sessions that correlated as strongly as
the aggregated scores did (mean correlation = .50, SD = .20).
Interestingly, context affected familiarity ratings differently for different stimuli:
there is a significant interaction of Context*Item (F(24, 2028) = 14.57, p <.001, η2 =.146).
Providing a context increases familiarity ratings, but only for items at the lower end of
the frequency scale. There was no difference between isolated and contextualized items
in the degree of individual variation in judgments.
91
To gain insight into the extent to which variation is related to the rating scale used,
we conducted a follow-up study. In this second study we compare the stability and
reliability of Magnitude Estimation judgments and Likert scale judgments.
We will discuss how both the stability and the instability we observed are real
reflections of language: a stable system in a speech community consisting of variable
and changing speakers. Our results suggest that judgment data are informative at
various levels of granularity. They call for more attention to individual variation and its
underlying dynamics.
References
Schütze, C.T. & Sprouse, J. (2013). Judgment Data. In D. Sharma & R. Podesva (Eds.), Research
Methods in Linguistics. Cambridge: Cambridge University Press.
92
A three-layer approach to the cognitive representation and
linguistic marking of subjectivity and perspective
Yipu Wei & Jacqueline Evers-Vermeul (Utrecht University)
Background: Connectives function as processing instructions for readers: they indicate
a type of coherence relation, and sometimes also mark a certain degree of subjectivity.
An example is Dutch want ‘because’, which marks a subjective causal relation, and
affects on-line processing (Canestrelli, Mak and Sanders, 2012). Processing studies
suggest that other linguistic elements such as epistemic modals (perhaps, probably) and
expressions with cognition/communication verbs (John thinks/says) also influence online processing and interact with the processing effects introduced by connectives
(Canestrelli et al., 2012, Traxler et al., 1997). These processing instructions have
previously been termed as perspective markers (Verhagen, 2005) or mental space
builders (Sanders, Sanders and Sweetser, 2012).
Research questions & Method: In this paper we address the following questions: What is
the role of linguistic elements such as connectives, epistemic modals and
cognition/communication verbs in representing perspective in discourse, and establish
specific degrees of subjectivity? How do these linguistic elements interact?
We will present our model and first show – on the basis of a literature study – how it
relates to notions from previous studies. Then we will show how a variety of
observations from previous studies on perspective and subjectivity can be accounted for
in a unified and integrated approach.
Analysis: We propose to apply a three-layer framework, based on Halliday’s (1985)
three metafunctions). According to this framework, information coded by linguistic
elements is functional at different layers: the propositional layer, the discourse layer
and the interpersonal layer. We will show that perspective markers can be categorized
according to the layers they function at. Furthermore, this three-layer model provides a
new viewpoint to understand subjectivity and perspective. Subjectivity in coherence
relations is formed at the discourse layer, while the extra cognitive effort brought about
by subjectivity is relieved at the interpersonal layer by the explicit marking of
perspective.
93
References
Canestrelli, A., Mak, W. & Sanders, T. (2013). Causal connectives in discourse processing: How
differences in subjectivity are reflected in eye-movements. Language and Cognitive
processes, 28(9), 1394-1413.
Halliday, M. (1985). An Introduction to Functional Grammar. Edward Arnold Publishers: London.
Sanders, T., Sanders, J., & Sweetser, E. (2012). Responsible subjects and discourse
causality. How mental spaces and perspective help identifying subjectivity in Dutch
backward causal connectives. Journal of Pragmatics, 40(2), 191-231.
Sanders, J., & Spooren, W. (1997). Perspective, subjectivity, and modality from a cognitive
linguistic point of view. In W. Liebert, G. Redeker & L. Waugh (Eds.), Discourse and
Perspective in Cognitive Linguistics (pp. 85–112). Benjamins: Amsterdam.
Stukker, N., & Sanders, T. (2011). Subjectivity and prototype structure in causal connectives: A
cross-linguistic perspective. Journal of Pragmatics, 40(2), 169-190.
Traxler, M., Sanford, A., Aked, J., & Moxey, L. (1997). Processing causal and diagnostic
statements in discourse. Journal of Experimental Psychology: Learning Memory, and
Cognition, 23(1), 88-101.
Verhagen, A. (2005). Constructions of intersubjectivity. Discourse, syntax, and cognition. Oxford:
Oxford University Press.
94